Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
The role of GIS in asset management: integration at the Otay Water District
(USC Thesis Other)
The role of GIS in asset management: integration at the Otay Water District
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
THE ROLE OF GIS IN ASSET MANAGEMENT:
INTEGRATION AT THE OTAY WATER DISTICT
by
Alexander John Schultz
A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(GEOGRAPHIC INFORMATION SCIENCE AND TECHNOLOGY)
December 2012
Copyright 2012 Alexander John Schultz
ii
ACKNOWLEDGEMENTS
During the development and writing of this thesis study, it became apparent to me that it
could not be accomplished alone and I needed the help of others. This project could not have been
completed if it were not for the access to data, people, software, and other resources granted to
me as an employee of the Otay Water District. I would like to thank Ming Zhao, GIS Manager,
and Nader AlAlem, GIS Contractor from Halax2 Inc., for their help in the development of the
asset database structure. I also would like to thank Dongxing Ma, GIS Programmer Analyst, for
creating the asset database entry form. Thanks to Jake Vaclavek, Water Systems Supervisor, Gary
Stalker, Water Systems Manager and Don Henderson, Water Systems Operator, for evaluating
my model results and giving me feedback. Also thanks to William Granger, Water Conservation
Manager, and William Poulin, Database Administrator, for assisting me in identifying high
consumption parcels. Lastly, I would thank my thesis advisor Dr. Robert Vos for his countless
hours of assistance and support. I would also like to thank the other thesis committee members,
Dr. Karen Kemp and Dr. John Wilson for their added input to my thesis study.
iii
TABLE OF CONTENTS
Acknowledgements ii
List of Tables v
List of Figures vii
List of Acronyms ix
Abstract x
Chapter 1: Introduction 1
1.1 GIS and Asset Management 3
1.2 GIS Implementation at Otay Water District 5
1.3 Study Objective at Otay Water District 7
Chapter 2: Background 8
2.1 Asset Management and Water Utilities 8
2.2 GIS and Asset Management 11
2.3 Computerized Maintenance Management Systems and Enterprise
Asset Management 13
2.4 Asset Risk, Condition and Criticality 16
Chapter 3: Methodology of Case Study of Otay Water District (OWD) 18
3.1 Scope of Study 18
3.2 Database Design 19
3.3 Populating the Database 25
3.4 Risk Factor Analysis 26
3.5 Spatial Parameters 27
3.5.1 Customers Served by Pump Station 28
3.5.2 Hydrants Served by Pump Station 29
3.5.3 Schools Served by Pump Station 30
3.5.4 Number of High Consumption Users 30
3.5.5 Presence of a Hospital 31
3.5.6 Serves a Major Municipality 31
3.5.7 Multiple Fire Stations 32
3.6 Non-Spatial Parameters for Criticality 32
3.6.1 Pump Redundancy Lost 33
3.6.2 Pump Station Redundancy 34
3.6.3 Reservoir Redundancy 34
3.6.4 Time to Restore Service 36
3.7 Probability Scoring 36
iv
3.8 Asset Risk Determination Using ModelBuilder 37
Chapter 4: Results 39
4.1 Detailed Results 41
4.2 Expert Confirmation of Model Results 47
4.3 Sensitivity Analysis 49
Chapter 5: Conclusions 58
5.1 Model Refinements 58
5.2 Benefits to Otay Water District 59
5.3 Otay Accomplishments Present and Future 60
5.4 Value of Integrating GIS with Asset Management in Utilities 62
References 63
Appendices 67
Appendix A Table displaying results for equal interval classification 67
Appendix B Integration of the spreadsheet into data model 70
Appendix C Asset Data Entry Form showing examples of parent assets 71
Appendix D Example of Asset Data Entry Form for pump station 72
Appendix E Feature class relating to abstract class 73
Appendix F Diagram of model used to determine risk factor 74
v
LIST OF TABLES
Table 1: Scoring range for nominal spatial parameters 28
Table 2: Scoring range for binomial or trinomial parameters 28
Table 3: Scoring range for customers served by pump station 29
Table 4: Scoring range for number of hydrants served 29
Table 5: Scoring range for number of schools 30
Table 6: Scoring range for high consumption users 31
Table 7: Scoring range for hospitals 31
Table 8: Scoring ranges for municipalities 32
Table 9: Scoring ranges for fire stations 32
Table 10: Scoring ranges for non-spatial features 32
Table 11: Scoring for pump redundancy lost 33
Table 12: Scoring for pump station redundancy 34
Table 13: Scoring for reservoir redundancy 35
Table 14: Scoring for time to restore service 36
Table 15: Scoring range for criticality 36
Table 16: Scoring range for probability 37
vi
Table 17: Scoring for Risk Factor for equal interval classification 40
Table 18: Risk factor results for Pump 1 of the Low Head pumps 42
Table 19: Risk factor results for 980-1 pumps 43
Table 20: Risk factor results for Pump 1 of the 711-1 pumps 45
Table 21: Risk factor results for 1090-1 pumps 46
Table 22: Risk factor results for 944-1 pumps 47
Table 23: Scoring range for risk factor using natural breaks (Jenks) 51
Table 24: Risk factor results for 944-1 pumps equal interval and
natural breaks (Jenks) 51
Table 25: Scoring for risk factor for redundancy x 10 53
Table 26: Risk factor results for 944-1 pumps increasing x 10 53
Table 27: Scoring for risk factor for pumps with mean criticality 54
Table 28: Risk factor results for 944-1 pumps with mean criticality 55
Table 29: Scoring range for risk factor with maximum criticality 56
Table 30: Risk factor results for 944-1 pumps with maximum criticality 57
vii
LIST OF FIGURES
Figure 1: Core Water Utility Business Patterns 12
Figure 2: Current operating pump stations in the Otay Water District 20
Figure 3: Pump station 711-1 21
Figure 4: Access database table showing asset account data structure 22
Figure 5: Example of feature classes inheriting abstract class attributes 23
Figure 6: Integration of GIS and asset management 23
Figure 7: GIS and asset management system with vertical assets 24
Figure 8: Detail of pump station data structure 26
Figure 9: Partial view of the Potable Hydraulic Profile Schematic 35
Figure 10: Results for risk factor using equal interval rankings 40
Figure 11: Chart displaying results for pumps that scored in each risk factor
category 41
Figure 12: Results for risk factor using natural breaks (Jenks) classification 50
Figure 13: Results for risk factor using natural breaks (Jenks) and 52
weighting redundancy x 10
Figure 14: Risk factor scoring results after populating 944-1 pumps' spatial 54
parameters with mean score
viii
Figure 15: Risk factor scoring results after populating 944-1 pumps' spatial 56
parameters with maximum score
Figure 16: Chart displaying different scoring results for 944-1 pumps 57
ix
LIST OF ACRONYMS
ASCE . . . . . . . .American Society of Civil Engineers
AWWA . . . . . . American Water Works Association
CIP. . . . . . . . . . Capital Improvement Project
CMMS. . . . . . . Computerized Maintenance Management Systems
EAM . . . . . . . . Enterprise Asset Management
EPA . . . . . . . . .U.S. Environmental Protection Agency
ESRI . . . . . . . . Environmental Systems Research Institute
GAO . . . . . . . . U.S. General Accounting Office
GASB . . . . . . . Governmental Accounting Standards Board
GIS . . . . . . . . . Geographic Information System
GPS . . . . . . . . .Global Positioning System
GUI . . . . . . . . . Graphic User Interface
IIMM. . . . . . . .International Infrastructure Management Manual
MCC. . . . . . . . Motor Control Cabinet
NAMS . . . . . . Australia and New Zealand's National Asset Management Steering
O&M. . . . . . . Operations and Maintenance
OWD . . . . . . . Otay Water District
RDMS. . . . . . . Relational Database Management System
SCADA . . . . . Supervisory Control and Data Acquisition System
SQL. . . . . . . . . Sequential Query Language
WIN . . . . . . . . Water Infrastructure Network
x
ABSTRACT
This study demonstrates the integration of Geographic Information Systems (GIS) with
asset management. There are few existing studies or demonstrations of the integration of GIS
technology with asset management systems, especially for vertical assets at water utilities. A
model is developed using Otay Water District (OWD) as a case study. The case study expands
upon a GIS model that already contains horizontal assets (e.g., pipelines). The new model
includes vertical assets (e.g., pump stations). In the past, non-spatial vertical assets, such as pump
stations and their components were represented only by a point and could not be plotted against
spatial data variables. In the expanded model, spatial and non-spatial asset risk variables are
measured and scored for the 79 pumps within the 20 pump stations at the district. Each pump is
assigned criticality and probability scores, which are then multiplied to give an overall risk factor
score. Model scores were plotted on a point symbology map and expert confirmation was
conducted with OWD water operations staff. A sensitivity analysis of the model reveals that
manipulating model parameters to increase overall scoring accuracy of some pumps can also have
a negative impact on the scoring of others. Further study is needed to plan and implement
schemes that allow vertical assets at utilities to inherit asset management scores based on their
positions within larger horizontal networks.
1
CHAPTER 1 – INTRODUCTION
All water utilities are made up of assets. The physical assets of a water distribution system
include pipelines, storage reservoirs, pump stations, hydrants, valves, meters, manholes, and any
other components that make up the system. Assets can be categorized as either horizontal or
vertical. Vertical assets are those that are primarily above the ground, such as pumps, reservoirs,
and treatment facilities. The horizontal assets are usually the buried assets such as the water
mains that form the backbone of the water distribution and wastewater collection systems (New
Mexico Environmental Finance Center, 2006). Assets can contain other assets. For example, a
pump station can house important assets such as motors and an electrical system that support the
pumps (Zhao and Stevens, 2011).
As the U.S. water distribution system ages and deteriorates, the assets of the system
generally lose value and costs of operation and maintenance increase. Asset management is
concerned with strategic approaches to optimize cost effectiveness with decisions that balance
new investment and maintenance activities. In 2008, the U.S. Environmental Protection Agency
(EPA) referred to asset management as maintaining a desired level of service for a given set of
assets at the lowest cycle cost (U.S. EPA, 2008a). Lowest cycle cost is the least cost for
rehabilitating, repairing, or replacing an asset over a given amount of time (U.S. EPA, 2008a).
For a water utility, the management of assets plays a significant role in overall financial
performance.
An effective asset management system must include an effective maintenance
management system which is focused on reducing the maintenance cost while extending the
useful life of the asset (Shamsi, 2005). Many utilities use a react- to-crisis management approach
in dealing with infrastructure problems. This is usually not the best approach given the additional
costs of emergency crews and property damage. With the use of effective asset management, it is
2
possible to reduce overall infrastructure costs instead of waiting until the assets fail incurring
higher than necessary costs (Shamsi, 2005).
This is especially important because at present, aging water and wastewater
infrastructures in the United States are in critical stages of deterioration requiring billions of
dollars for renovation (ASCE, 2009). Many systems are not getting the necessary maintenance
and repairs needed to keep them working properly because of insufficient funding. In 2009, the
American Society of Civil Engineers (ASCE) released its annual report card for America's
infrastructure in which the nation's wastewater and drinking water systems each received a grade
of D minus. The ASCE reported that U.S. water systems have at least an $11-billion annual
investment shortfall to replace aging facilities and comply with existing and future federal safe
drinking water regulations. The shortfall does not account for any growth in drinking water
demand in the next 20 years (ASCE, 2009).
According to the American Water Works Association (AWWA), the cost of repairing and
expanding the United States drinking and water infrastructure will total more than $1 trillion
between 2011 and 2035 and exceed $1.7 trillion by 2050. The need will double from about $13
billion a year today (2012) to almost $30 billion (in 2010 dollars) annually by the 2040's
(AWWA, 2012). The $1 trillion estimate covers buried drinking water assets only. Above ground
drinking water facilities such as storage tanks, reservoirs and treatment plants will add to the
total.
In 2001, the Water Infrastructure Network (WIN), a consortium of industry, municipal
and non-profit associations, reported that the use of innovations in technology and management
by utility companies has cut operations and maintenance costs by 15% to 40% (WIN, 2001). One
of these innovative technologies is GIS which helps to analyze and communicate geographic or
spatial information associated with physical assets. According to Shamsi (2005), except for the
3
computer itself, no technology has so revolutionized the water industry as GIS. Another
innovation is a Computerized Maintenance Management System (CMMS). It can be implemented
for the more efficient maintenance of a utility because it accurately tracks problems within the
utility network. GIS and CMMS integration can facilitate proactive (preventative) maintenance.
Global Positioning System (GPS) is a key technology because it is used to increase the accuracy
of existing system maps by verifying and correcting locations of system components. Also maps
for new water systems can be created if they do not exist and water system attributes can be
collected for populating a GIS databases.
Along with budgetary constraints, there are increased governmental requirements that
affect the management of water utilities. For example, Rule 34 of the Governmental Accounting
Standards Board (GASB), requires cities to adequately account for and report their capital asset
inventory in a complete, accurate, and detailed manner. Because of the higher standards, GASB
Rule 34 is an important factor toward improved asset management. Congress has been
considering making utilities develop comprehensive plans as a condition for future funding
(GAO, 2004).
1.1 - GIS and Asset Management
GIS had been proven to be an effective and powerful tool in the water distribution industry.
According to the AWWA, as of 2002, 90% of water agencies were at least partially using GIS to
assist in applications (Shamsi, 2005). An application is an applied use of technology which
bridges the gap between pure science and applied use. An example for use in the water utility is a
CMMS. It can have many functions. For example, it can provide maintenance cost and history
along with providing asset inspection data and asset condition assessment. Integrating with a GIS
can improve the capabilities of a CMMS by supporting spatial analysis and locating
geographically dispersed facilities in the water system. A GIS is a special type of information
4
system in which the database of spatially distributed features and procedures collect, store,
retrieve, analyze, and display geographic data. GIS relates database records and associated
attribute data to a physical location, creating a "smart map" (Vanier, 2004). A GIS is also a means
of effectively analyzing large amounts of spatially related data. Making informed infrastructure
maintenance decisions requires large amounts of diverse information on a continuing basis. GIS
integrates all kinds of information from disparate sources into one manageable system so better
and informed decisions can be based on all relevant factors.
With the integration of information from a variety of sources, it is possible to
determine important geospatial relationships and factors on which utility maintenance would be
based. For example, water main failure could be caused not only by age, but also by pipe
material, surrounding soil, water pressure, and street traffic. By analyzing these factors and other
related factors, it would be possible to determine which assets are the "hot spot" areas and
constitute a priority for maintenance activities.
According to Shamsi (2002), the use of GIS technology can be an ideal solution for the
effective management of water industry infrastructure because it offers the power of both
geography and information systems. The key element of information used by a water utility its
location to geographic features and objects. According to some estimates, more than 80% of all
information used by water utilities is georeferenced making GIS technology especially applicable
as a management tool (Shamsi, 2002).
Spatial location is typically a major common aspect of all the data at a water utility. A
GIS can locate the exact position of a utilities infrastructure such as valves, hydrants, meters,
pumps, and manhole covers displaying them on a computerized map. It can also store important
data about each asset, including manufacturer, year of installation, repair history, size volume,
water quality data or almost any other type of information. Efficient management must include
5
location information so good decisions can be made relative to the surrounding area and affected
assets. With the use of GIS in the area of asset management, it is possible to visualize and
understand the geographical context of an asset and improve the efficiency of asset management.
1.2 - GIS Implementation at Otay Water District
The Otay Water District began implementing its GIS in 1995 and wanted to make the most of its
investment by fully realizing the potential of GIS (Zhao and Stevens, 2002). Major data needed
by the district were collected. The database was significant in size capturing the major attributes
of the facility infrastructure such as diameter, material, as-built number, facility page number, etc.
For its business operations, the district needed to keep a complete and detailed inventory,
including location and condition of all assets. GIS has been shown to be a state-of-the-art
technology which can efficiently perform the district's data related processes (Zhao and Stevens,
2003).
An Arc Internet Map Server (ArcIMS) based GIS web application was developed which
could be used by the District staff through the internet. A customized ArcIMS application was
developed with a similar Graphic User Interface (GUI) of ArcView desktop providing consistent
user interface for field laptop and other desktop applications. The interactive maps allow users to
query the data to derive more information. Also the web portal is a cost efficient way to distribute
geographic information to the GIS user. Before the implementation of the GIS, obtaining records
involved physically going into the record room and manually searching for the needed
information. This process was inefficient, error-prone and hindered the productivity of the water
district (Zhao and Stevens, 2003).
While collecting GIS data over several years, the district saw the need to integrate a
variety of information and applications with a geographic component into one manageable
system. The focus of GIS became one of a centralized asset for sharing and managing information
6
rather than a cartographic tool. The result was the District's enterprise solution which was
implemented in 2002 (Zhao and Stevens, 2002).
Enterprise GIS is an organizational approach that integrates various departmental projects
into a centralized GIS which serves as a foundation in integrating other tabular database systems
within the district. The core of the integrated systems for the district relates to customers,
financial management, work management, and GIS. Also included are important systems of
Supervisory Control And Data Acquisition (SCADA) for fuel and plant specific systems. SCADA
is a computer system for gathering and analyzing real time data. In 2003, the district adopted
Lucity as its CMMS mainly as the work order management tool. It is still in use. (Zhao and
Stevens, 2009).
A key component in the enterprise GIS is the database design. The district used Esri's
water utilities data model as the prototype to design the enterprise GIS database--including the
potable water, recycled water, wastewater and land-based systems. The (Structured Query
Language) SQL-based Geodatabase served as the basis for the district-wide enterprise system
integration. The open platform of this database structure made it possible for the district's GIS
system to integrate with other systems. Between-system integration is essential to make the most
of enterprise GIS (Zhao and Stevens, 2009).
During 2007 and 2008, the district reevaluated the GIS architecture including hardware,
servers, storage, network, applications from different systems, database requirements and user
requirements (Zhao and Stevens, 2009). The current system architecture is now designed to
accomplish the district's goal of higher availability, and better performance with current and
future enterprise integration.
Compared with the GIS technology capability of a decade ago, new GIS technology is
enterprise enabled and the district is headed in that direction (Zhao and Stevens, 2009). With the
7
enterprise approach, all operational data should be available and integrated. As part of Otay
Water District's strategic plan, the district wants to leverage its GIS investment with an enterprise
integration strategy (Zhao and Stevens, 2009). This includes a GIS-centric management system
which expands the existing water model of horizontal assets to integrate vertical assets. This
initiative is the inspiration for this thesis study which was undertaken with the cooperation of the
district.
1.3 - Study Objective at Otay Water District
The purpose of this study is to demonstrate and evaluate the integration of GIS and utility asset
management in Otay Water District. The district, located in the southern part of San Diego
County is the second largest in the county encompassing 125 square miles and serves the water
and/or sewer needs of a population of approximately 206,000 (OWD, 2011). The project involves
the expansion of the district's asset management system, composed of horizontal assets, to include
its vertical assets in detail. Even though the majority of the district's water utility assets are
horizontal, the vertical assets can be over 50% of the district's capital expenditures in cost
maintenance, repair, or replacement (Zhao and Stevens, 2011). It is important to develop ways to
try to economize with these expensive assets. One of the district's GIS strategic plan objectives is
to develop and implement an asset management program plan to extend the useful life of the
capital assets (Otay Water District, 2008). Another objective is to develop and test a criticality
analysis (composed of measures of consequence and risk of failures) for the 79 potable pumps
within the district's 20 pump stations.
8
CHAPTER 2 – BACKGROUND
Literature in the area of asset management in water utilities comes from a variety of sources,
including government publications, trade magazines, and conference proceedings. The
development of asset management approaches with water utilities originated around 2000.
Literature relevant to providing background for this study includes both an overview of work
done in asset management with water utilities and the narrower topic of the integration of GIS
with asset maintenance management--specifically CMMS. This section also includes the reviews
of the few articles found concerning the use of vertical assets, CMMS, and asset risk--the focus of
the thesis study.
2.1 - Asset Management and Water Utilities
The use of asset management in water utilities is a relatively new concept. Until around 2000, it
was relatively unknown in North America (Lutchman, 2006). The term originally described the
management of financial assets. In the past decade, an interest in asset management for water and
water utilities has grown mainly due in part to an aging water utility infrastructure. Many
professional and government organizations have defined asset management and developed plans
for the practice and implementation of asset management in the area of water utilities (Sinha,
n.d.).
In 2007, the U.S. Environmental Protection Agency (EPA) and six national water and
wastewater associations collaborated on a guide promoting effective utility management. The
guide discusses ten attributes of effectively managed water utilities. It concludes that effective
asset management can enhance the infrastructure, improve performance in many critical areas,
and respond to current and future challenges (EPA, 2008b). The EPA also works with water
utilities to provide technical assistance to help utilities implement asset management (EPA,
2008a).
9
A report by the United States General Accounting Office (GAO) discusses the benefits of
comprehensive asset management for drinking water and wastewater utilities. It also addresses
the challenges of implementation and the federal government's role in encouraging utilities to use
it (GAO, 2004). Utilities reviewed by the GAO reported that collecting accurate data about their
assets in areas like maintenance, rehabilitation, and replacement costs can lead to better
investment decisions. The challenges include collecting and managing needed data and
integrating information and decision making across departments. Also, it is reported that the
shorter-term focus of those in charge of utilities can hamper long-term planning efforts. The
federal government has invested billions of dollars in drinking water and wastewater
infrastructure and wants to protect its investment by having future funds go to those utilities
which implement comprehensive asset management plans (GAO, 2004).
One of the goals of asset management is to replace reactive maintenance with planned
maintenance with more practices geared toward predictive and condition maintenance (Harlow,
2000, part 1). Good asset management must minimize long-term asset costs and at the same time
insure reliable customer service. Effective asset management must be based on practices that are
easily implemented, cost effective, and sustainable in the long run (Lutchman, 2006). Lutchman
(2006) also believes that good asset management needs to be focused on economic, social and
environmental concepts and not just the financial bottom line.
A typical asset management framework consists of the following four parts: 1.) Facilities
inventory; 2.) Condition assessment; 3.) Operations, maintenance, repair and replacement
management; 4.) Analysis and evaluation (Doyle and Rose, 2001). Facilities inventory is a
description of each asset. Condition assessment classifies each asset as to its capability to perform
its intended function. That function being operations, maintenance, repair and replacement
management tracks and records data about work orders and customer complaints. It also issues
10
and tracks preventative and predictive maintenance schedules, generating crew assignments and
work-site maps. Analysis and evaluation prioritizes work effort, analyzes cost effectiveness, and
optimizes asset performance.
Asset management was first done by the water, wastewater and public works utilities in
New Zealand and Australia. They set the general direction and standards for asset management in
these industries. In the mid 1980's and 1990's, the government directed the utilities to become
business based, customer focused, more transparent and accountable. Policies and regulations
were set and the utilities were mandated to meet them. During a 12 year period, the 24 largest
Australian water and wastewater utilities achieved almost a 20 percent savings per customer
account. Savings involved capital and operations/maintenance costs with no changes in service
levels to the customer. A large regional and wastewater utility (Hunter Water located in New
South Wales, Australia) achieved far more savings and is widely viewed as having developed one
of the most effective and advanced asset management programs in the world (Sinha, n.d.).
Australia and New Zealand’s National Asset Management Steering (NAMS) group have
developed an international infrastructure management manual which is considered to be one of
the best sources for public utility asset management information (Harlow, 2000, part 4). It was
first introduced in 2000, followed by a number of revisions including the latest 4th edition of the
manual (NAMS, 2011). The manual includes five sections beginning with an introduction to the
concepts of total asset management and lifecycle asset management. The other sections include:
implementing asset management, implementation techniques, asset management information
system and data management. There is also country specific information with best practices for
not only Australia and New Zealand, but also other countries such as the United States and
Canada (Sinha, n.d.).
11
The Seattle Public Utilities imported the asset management concept in the early 2000s.
They became one of the first in the United States to implement a formal asset management
program. Computerized asset management systems can cost $1million to $2 million for a large
utility. But the Seattle utility company estimates its computerized asset management system has
saved the city more than $180 billion (AWI, 2010).
2.2 - GIS and Asset Management
Even though GIS technology began in the 1960s, GIS applications for the water industry did not
evolve until the late 1980s. In the early 1990s, the water industry started to use GIS in mapping,
modeling, facilities management and work- order management plans. By the end of 2000,
approximately 90% of the water utilities in the United States were using GIS technology in some
form (Shamsi, 2005). The use of GIS as a management tool has grown since the 20th century and
the number of users has increased substantially. Utilities that are using GIS successfully have
seen increased productivity and increased efficiency which saves time and money. The
Environmental Systems Research Institute (Esri), the leading GIS software company in the world,
has been a significant contributor to GIS applications in the water industry. In 2009, Esri started a
Water Utility Resource Center for the utility needs of over 300,000 worldwide users (Baird,
2011). The website is: (http://resources.arcgis.com/content/water-utilities).
As shown in Figure 1 from Esri, asset management is one of the core business patterns
commonly used by water utilities. Others include: planning and analysis, field mobility,
operational awareness and stakeholder engagement (Crothers, 2011). The use of GIS can be an
important part of each of these patterns.
12
Figure 1 - Core Water Utility Business Patterns.
Source: Crothers, 2011. Esri.
The basis of effective asset management at a water utility is good asset information. GIS
manages asset information by storing, managing, and maintaining accurate asset records that can
be shared by the whole utility. Many times, a water utility will have complete information about
an asset stored in multiple systems. The GIS stores the location, connectivity to other assets and
basic attributes. The CMMS stores extended information about the work history for an asset.
Other systems could include a financial system and a customer information system.
There should be integration among all of the multiple systems that store information
about an asset so data about its location, connectivity, status, history and description can be easily
accessed. A GIS has information that can be shared across an entire utility and used to support
many of its information needs. Utilities can significantly increase their return in a GIS investment
by sharing it around the entire utility and using it to support its many business patterns (Crothers,
2010).
Asset data in a GIS can be used to support the planning needs of a utility through spatial
analysis. Water utilities are involved in short term planning and long term planning. For short
term planning, GIS is used in creating and optimizing reactive and proactive work orders. Long
13
term planning involves the use of asset data, performance data and GIS analysis to understand
how an individual utility is performing. This information is used to help determine where to best
spend capital funds to maximize the value of a utility's assets (Crothers, 2011).
GIS supports the field mobility business pattern by providing field crews with maps and
applications that can be rapidly updated and are easy to use. It also enables field crews to capture
GIS data and send it back to the central office. The operational awareness business pattern
involves the performance of assets, utility networks and personnel and how they are affecting
each other. Utility managers can then make decisions based on accurate and up to date
information. GIS supports this by enabling utilities to have an interactive map of the current state
of operations. An interactive map is an easy way to take information from many systems and
present it through a common application (Crothers, 2011).
The final business pattern, stakeholder engagement, involves sharing information with
stakeholders such as customers, elected officials, regulatory agencies, and other utilities in the
service area. The trend is for water utilities to actively engage with stakeholders through public
outreach programs providing accurate information that minimizes misinterpretation. GIS is used
by utilities by creating static and interactive maps. Mapping applications for stakeholders include
customer self service, capital project coordination, service interruption management and
transparency into utility performance (Crothers, 2011).
2.3 - Computerized Maintenance Management Systems and Enterprise Asset Management
Enterprise Asset Management (EAM) Systems and CMMS are being implemented by a growing
number of water and wastewater utilities and their use in these areas appears to be growing. A
CMMS is a software package which maintains a computer database about a utility's maintenance
operations. The terms asset management and maintenance management are often used
interchangeably. Even though they are related, they are different processes with different
14
objectives. Asset management is focused on reducing the maintenance cost of ownership, while
maintenance management is focused on reducing the maintenance cost while extending the useful
life of an asset (McKibben and Davis, 2002). An effective asset management system must include
an effective maintenance management system and is considered the most important core of an
asset management system (Shamsi, 2005).
An EAM is a CMMS focused on maintenance work orders and performance combined
with an asset registry or inventory. A CMMS and asset registry are the center of an asset
management program. A growing trend is for the GIS geodatabase to be the starting point for an
asset management program and the asset inventory. Utilities need to know the location and
condition of their assets. GIS is the best place for gathering asset data because spatial location is
typically an important aspect of the data at a water utility. The GIS geodatabase combined with
the CMMS forms a comprehensive customer request, asset inventory and work management
system and becomes the foundation for the EAM. This combination captures asset data, work
history and condition assessments necessary for cost-effective, condition and predictive
maintenance programs (Baird, 2011).
McKibben and Davis (2002) give many reasons for the integration of GIS and CMMS.
GIS can significantly enhance a CMMS by providing the ability to access, use, display, and
manage spatial data. This is important for utilities with geographically dispersed networks. Also
it can provide access to other spatial data. It can provide maps of the utility that can be used in
locating facilities included in work orders. It can be used to effectively schedule and assign
maintenance work crews to certain work locations, saving time and cost. Baird (2011) points out
that using a GIS with full functionality, beyond just map-making, will result in lower
maintenance costs and a lower cost EAM.
15
At the time of their study, McKibben and Davis (2002) found there were only six CMMS
vendors that had links to ESRI GIS software. Of these, only three, Azteca Systems (Cityworks),
GBA Master Series, and Hansen Information Technologies (Hansen's Citizen Relationship
Software) had useful CMMS systems for water and wastewater utilities.
There are two methods of integrating GIS with the CMMS and these are based on where
the asset data is stored--the CMMS or the GIS. GBA and Hansen maintain the asset data in the
CMMS database and GIS software is used to access the asset data or provide information stored
in the CMMS. GIS features are linked to the CMMS database. Adding a new asset requires the
addition of the asset to both the GIS and CMMS databases. The work order and maintenance data
is stored in the CMMS (McKibben and Davis, 2002).
Azteca's Cityworks uses the other method which stores the asset data in the GIS database.
All assets and the related data are maintained in the GIS database. The addition of new assets in
the GIS database does not require an adjustment to the Citywork's database. Work orders and
maintenance management functions are maintained in a series of Cityworks tables and all of the
maintenance management functions are provided as extensions of ESRI's GIS software
(McKibben and Davis, 2002). Azteca's Cityworks, with its GIS-centric approach, is considered
by many to be one of the best asset and maintenance management systems (Baird, 2011). It has
been in use for 15 years and has over 400 clients. According to Baird (2011), a solid CMMS is a
necessary part of asset management. Therefore, a CMMS with a GIS-centric approach is
considered to be a necessary part of asset management.
As previously mentioned in the introduction, water utilities own two major types of
assets--horizontal and vertical. Horizontal assets are geographically dispersed in the distribution
system and vertical assets are concentrated in a pump station or water treatment plant. Pump
stations and water treatment plants have a much larger number of assets and maintenance
16
activities than those of distribution systems (McKibben and Davis, 2002). McKibben and Davis
2002 stated that the use of GIS data within vertical assets may be beneficial, but requires more
research and development. The integration of GIS and Azteca's City Works CMMS eliminates
the need to implement two maintenance management systems because it does not require its own
database repository. Instead it directly accesses the asset management geodatabase.
Few studies exist on expanding horizontal asset management for CMMS-GIS into
vertical assets. Zhao and Stevens (2011) discuss expanding the Otay District's water model of
horizontal assets to integrate the districts vertical assets within a pump station. They also state
that with the present state of GIS software and technology, GIS has not been used to capture the
information within pump stations and treatment plants graphically. The model development and
assessment in this study contributes to the literature in this area.
2.4 - Asset Risk, Condition and Criticality
The concepts of asset condition, criticality and risk are important in the area of water utility asset
management. Asset risk is based on condition (probability of failure) and criticality (consequence
of failure). Two articles were found relating these concepts to vertical assets with the use of water
utilities. Hyer (2010) discusses the implementation of a pilot project at Florida's Toho Water
Authority demonstrating the collection and calculation of asset data. The data included
information on asset condition, consequence of failure, risk of failure and replacement cost for all
vertical assets in the water utility. The data analysis provided information for future renewal and
replacement costs based on asset risk and remaining useful life. With the model, the rest of the
vertical assets could be evaluated in a systematic way utilizing information for capital and O&M
budgeting and planning.
Hyer (2011) discusses in detail the use of condition, consequence of failure and risk
scoring with the building of a comprehensive asset management program for the Austin Water
17
Utility in Texas. The pilot project began with a vertical pilot asset inventory, determination of the
asset hierarchy and required asset attributes. Vertical asset condition scoring standards, criticality
scoring standards and risk-calculation scoring are also included. The asset condition assessment
scoring evaluated all aspects of asset failure to establish its probability. The assessment provided
information to determine specific short-and long-term capital needs based on asset risk and
remaining useful life. By establishing a specific scoring system, future condition assessments can
be performed consistently. By completing the pilot program, the rest of the vertical assets can be
evaluated in a systematic way. The scoring criteria and guides that were developed can be
adapted for other water utilities with simple revisions.
The two articles did not specifically discuss spatial factors in determining an asset's
criticality. An example would be customers that have water distributed to them by a pump station.
A customer is represented by the water meter which is located where the customer resides. This
would also be the same for other spatial variables such as fire hydrants, schools, hospitals, fire
stations that have water distributed to them by a pump station.
Also the articles were not specific as to how the vertical assets were represented in GIS.
They could be represented as having a spatial location or represented by a non-spatial table that
relates to a spatial feature. An example of the latter would be a pump data table with no spatial
location. The pump data table can be represented by pump station using a point data to a given
location. However since the pump station is only represented by a point, the service area of the
pump would not be able to be plotted correctly without further data. The non-spatial data tables
on vertical assets need to be related to spatial data with close coordination of data systems. This
thesis develops and reports on such data integration for the Otay Water District.
18
CHAPTER 3 - METHODOLGY OF CASE STUDY OF OTAY WATER DISTRICT (OWD)
The purpose of the case study is to demonstrate the integration of GIS with asset management in
a water utility. The project involves the expansion of OWDs current asset maintenance
management system of horizontal assets to include its vertical assets in detail. Currently, the
district's vertical assets, including pump stations, reservoirs, and a single treatment plant, are
represented as a point feature in the GIS. Developing the database for spatial attributes associated
with vertical assets is necessary to perform criticality analysis and score for asset risk.
Specifically, the objective is to develop a criticality analysis for use with the pumps within the
pump stations in order to determine proactive maintenance and replacement schedules.
3.1 - Scope of Study
As a pilot study for vertical assets, the study focuses on all the pumps within the district's potable
pump stations. Figure 2 provides a visual orientation for all the potable pump stations. Pump
stations help transport water through the mains located at different elevations through increased
pressure. Most of the district's pump stations are located in the hilly elevations in the northern
part of the district. Some of the district's pump stations are considered more pertinent than others
because they directly pump water to other pump stations. If the pump stations that provide water
to others were to fail then the pump stations that rely on other pump stations for water would be
greatly affected.
There are a variety of spatial factors that raise or lower the criticality of a pump or pump
station in the network. These include whether they serve hospitals, have more customers, contain
more fire infrastructure, and whether they serve many high consumption users. Another criticality
factor for pump stations are those that serve water within a municipality. Major municipalities
tend to have a higher population, more businesses and contain more infrastructure than pump
stations that serve unincorporated regions of the county.
19
There are also non-spatial parameters that raise or lower the criticality of a pump or pump
station. These include pump redundancy, and the estimated time to restore a pump service after
failure or maintenance has occurred. The objective of this study is to explore whether spatial
factors can assist in determinations regarding which pumps need to be replaced, and which pumps
take precedence above others when failure or replacement occurs at the same time.
3.2 - Database Design
The design of the database structure is very important in implementing this project. Relationships
must be established correctly between spatial feature classes and non-spatial feature classes
maintained by the district. When data are plotted correctly, classes of both spatial and non-spatial
features will be accessible in GIS. This combined horizontal and vertical data structure provides
for a GIS-centric management system. This strategy combines a CMMS with a GIS geodatabase,
creating the foundation for an EAM approach. The district's strategic plan emphasizes the
enterprise approach to GIS with the interdepartmental sharing of data meeting the needs of many
departments. The district's ultimate goal is cost effective business processes in the managing of
infrastructure assets (Zhao and Stevens, 2009).
A key component in the enterprise GIS is the database design. The district used Esri
water utilities data model as the prototype to design the enterprise GIS database--including the
potable water, recycle water, sewer collections and land-based systems (Esri, 2011). A series of
interviews were conducted among the different departments to make sure the design would fit the
end users' requirements. The SQL-based geodatabase serves as the basis for district-wide
enterprise system integration. The open platform of this database structure makes it possible for
the district's GIS system to integrate with other systems e.g., Azteca’s City Works and Riva’s
Modeling Software. Such integration is essential to make the most of enterprise GIS (Zhao and
Stevens, 2009).
20
Figure 2 - Current operating pump stations in the Otay Water District. Source: Alexander
Schultz, Otay Water District (2012).
Currently, pump stations along with reservoirs and the treatment plant are
represented as a point feature in GIS. Figure 3 displays an orthophoto of a pump station. Prior to
the database development in this study, selecting the pump station would result in the attributes of
the pump station only without any of the vertical assets contained within it. With the present state
of GIS software and technology, GIS has not been used extensively to capture the information
within pump stations (Zhao and Stevens, 2011). Even though the pump station is represented by a
21
point feature, in reality, the pump station contains other important assets that are not currently
represented in GIS such as motors, vessels, and an electrical system. The electrical system then
may consist of one or many other assets like a transformer and a Motor Control Cabinet (MCC)
board.
Figure 3 – Pump Station 711-1. Source: Otay Water District (2010).
An engineering consulting firm was contracted by the district to conduct a baseline asset
management assessment. The study reviewed the existing district data, general preventative
maintenance practices and established a fixed asset hierarchy schedule. The information for this
study was primarily taken from the district's infrastructure management system (IMS). It was put
into an Access spreadsheet format. The following Figure 4 is part of the asset account detail
structure.
22
Figure 4 - Access database table showing asset account data structure. Source: Otay Water
District (2011)
The database design was created in Visio by Nader AlAlem, GIS Contractor from Halax2
INC. and Ming Zhao, Otay Water District’s GIS Manager and later exported to Universal
Modeling Language (UML). The database structure was imported into an empty geodatabase.
Using the account detail spreadsheet and integration concepts, Nader AlAlem and Ming Zhao
created a UML diagram that was finally integrated with the existing Esri GIS data water model.
An important aspect of the database development is the relationship between the abstract
classes and the other asset feature classes. Abstract classes are those that cannot be instantiated
and cannot be used on their own but must be first inherited by a feature class. When a feature
23
class is linked to the abstract classes that feature class inherits the abstract class attributes. In the
model, multiple feature classes are sharing the same attribute classes. Install date, facility id, as-
built, and set id are attributes all used by feature classes pipes, valves and meters. Instead of
creating an attribute for each feature class, an abstract class is created that contains these
attributes. An example of this is shown in Figure 5.
Feature Classes Abstract Class attributes
Figure 5 – Example of feature classes inheriting abstract class attributes. Source:
Alexander Schultz (2012)
Figure 6 shows the basic idea of how GIS feature classes such as pump stations and
system valves would have corresponding tables in the asset management system. The GIS data
base relates to the asset management system on a one to one relationship. This is accomplished by
joining the fields GlobalID that are shared between the GIS and Asset Management System
(AMS).
Figure 6 - Integration of GIS and asset management. Source: Otay Water District (2011)
Feature Classes Attributes in abstract class
Pipes
Valves
Meters
Install Date
Facility ID
As-built
Set ID
Abstract
Class
24
Figure 7 shows the integration of the GIS feature class pump station with its
corresponding asset management system data. It also shows the asset hierarchy for pump station
contained in the AMS. The asset management table was created in the same GIS relational
database management system (RDBMS) using RDBMS Sequential Query Language (SQL)
statements. The new table was then registered with the geodatabase in ArcCatalog. Next,
relationships were established between these assets and the assets within each pump station. The
GIS feature class and the asset management table are related through a common Global ID. This
process was repeated to cover all the GIS feature classes.
Figure 7 - GIS and asset management system with vertical assets. Source: Otay Water District
(2011)
After these steps the vertical asset data is successfully integrated into the GIS data
structure. To test the success of the integration, when the pump station is selected using the
25
identify tool, the system will display the table showing GIS pump station with the asset database
features contained within it.
3.3 - Populating the Database
Work crews from the OWD Operations Department initially collected the vertical asset data
while in the field. Excel spreadsheets were printed to record the data. After the crews returned
from the field, the data were input into the asset management database. Before the pump station
data were fully populated, GIS programmer Dongxing Ma developed an asset management data
entry form using Visual Basic.Net technology. The entry forms provided multiple levels
hierarchies of an individual facility for the field staff to enter the asset information.
Following that, I populated the non-spatial data tables for this study by opening the asset
management data entry form installed on a desktop computer. Once the form was opened, a
parent asset facility (pump station, reservoir, treatment plant etc.) for the vertical asset was
opened (Refer to Appendix C). Then the specific facility for the vertical asset was selected from a
drop down menu. The tab for the vertical facility that needed to be added to the database was
selected (Refer to Appendix D).
By repeating this process for each vertical asset, the assets' attributes were populated into
the form. The new data were pushed over into the AMS. Selecting the pump station feature class
with the editor tool and opening the attribute table on the editor tool bar then allow us to view the
newly populated data.
Figure 8 demonstrates the data structure for a single pump station with three pumps, three
control valves and two engines. It also has a single electrical system, which consists of one
emergency power supply and two MCC boards. This further subdivision of the assets in the pump
station is shown in Appendix B. This illustrates the need to structure the integration of the GIS
26
and the asset databases carefully so that all the vertical assets within a given point feature will be
related to the correct asset feature tables.
Figure 8 - Detail of pump station data structure. Source: Otay Water District (2011)
3.4 - Risk Factor Analysis
The core method in this case study uses OWD data and SanGIS data to determine asset risk
factors for both spatial and non-spatial variables. To calculate risk factors, a model was
constructed in ArcGIS ModelBuilder to generate results for individual pumps that are housed in
pump stations throughout the district. After the results were calculated, a field study was
conducted to confirm results for a few pump stations in consultation with experts at OWD.
27
The two parameters that make up the asset risk factors are probability and criticality.
Criticality is defined as the consequence of asset failure or the severity of the impact. This is
based on an expert estimation of the consequences of failure-- who or what is affected if an asset
fails. There are both spatial and non-spatial aspects to criticality, including the redundancy of
pumps in the system. Probability is defined as the likelihood that a given asset (i.e., a pump) will
fail in a given length of time. In this study, probability is measured by the condition and age of an
asset. The overall risk factor is determined by multiplying asset criticality and the probability of
asset failure.
Criticality measures the consequence of failure. It is based on an assessment of how the
failure of a particular asset will affect a utility’s ability to meet its service goals. Criticality is used
to assist in prioritizing repair/replacement decisions, condition assessment and maintenance
activities. In this case study, the formula for the total criticality score is the addition of all spatial
and non-spatial parameters' scores following Hyer (2011).
3.5 - Spatial Parameters
The spatial parameters for criticality include: customers served by a pump station (number of
water meters), number of fire hydrants served by a pump station, number of schools, and number
of high consumption users. There are also parameters for the presence or absence of fire stations,
hospitals, and major municipalities in a pump zone. These are ranked numerically. A higher
number means a pump is more critical and a lower number means less critical. The spatial
parameters that are ordinal variables and scoring ranges are displayed in Table 1. These include:
customers served by a pump station, number of hydrants served by a pump station, number of
schools, and number of high consumption users. The spatial parameters that are binomial or
trinomial are displayed in Table 2. These include: fire stations, major municipalities, and
hospitals.
28
The spatial parameters are determined by spatially joining to the pressure zone layer.
The pressure zone layer is used because each pressure zone represents an area served by a pump
station. After the layers have been spatially joined, the frequency for each feature class is
generated. The features for each feature class are then ranked from 1 to 10 using equal intervals
for the nominal variables in Table 1. The equal interval classification is used so the attributes will
be grouped into an equal range of values. If there is a hospital or major municipality in the pump
zone, criticality will score a 10. The one spatial variable that is trinomial, fire stations is displayed
in Table 2. Fire stations are scored 0 if there are no fire stations present, 5 if one fire station is
present, and 10 if two or more are present. The hypothetical minimum score for spatial
parameters of criticality is 4 and the hypothetical maximum score for criticality is 70.
Table 1 – Scoring range for nominal spatial parameters.
Criticality
factor
Customers
served by
pump station
Number of
hydrants
served by
pump station
Number
of schools
Number of high
consumption
users
Ranking 1 - 10 1- 10 1 - 10 1 - 10
Table 2 – Scoring range for binomial or trinomial parameters.
Criticality
factor
Has a
hospital
Is in a major
municipality
Multiple Fire
Stations
Ranking 0 or 10 0 or 10 0 or 5 or 10
3.5.1 - Customers served by pump station
The data that were used to determine the customers served by pump station comes from Otay’s
meter feature class. The number of customers served by a pump station is measured by the
number of meter services found in each pump stations pump zone. A higher number of meter
services mean greater criticality of the pump within the pump station. For example, to find the
29
customers served by pump station 711, 711 pressure zone is used because that is the zone that the
pump station serves. Table 3 shows the scoring range for customers served.
Table 3 – Scoring range for customers
served by pump station served.
Data Range Score
2 - 1264 1
1264 - 2526 2
2526 - 3788 3
3788 - 5050 4
5050 - 6312 5
6312 - 7574 6
7574 - 8836 7
8836 - 10098 8
10098 - 11360 9
11360 - 12623 10
3.5.2 - Hydrants served by pump station
The data to determine the number of hydrants found within each pressure zone come from Otay’s
hydrant layer. The zone that contains the most hydrants is found to be the most critical. The
hydrant's scoring range is found in Table 4.
Table 4 – Scoring range for number
of hydrants served.
Data Range Score
4 - 118 1
118 - 234 2
234 - 349 3
349 - 464 4
464 - 580 5
580 - 695 6
695 - 810 7
810 - 925 8
925 - 1040 9
1040 - 1156 10
30
3.5.3 - Schools served by pump station
The data to measure the number of schools come from joining the SanGIS’ schools layer to
Otay’s pressure zone layer. The pressure zone with the highest number of schools is the most
critical. Equal interval is used to classify the pumps by number of users per each zone. Rankings
are based on a 1 to 10 scale. The schools' scoring range is in Table 5.
Table 5 – Scoring range for number
of schools.
Data Range Score
1 - 2 1
2 - 3 2
3 - 4 3
4 - 5 4
5 - 7 5
7 - 8 6
8 - 9 7
9 - 10 8
10 - 11 9
11 - 13 10
3.5.4 - Number of High Consumption Users
Users that eclipse the 100,000 cubic gallons per year threshold are considered high consumption
users. Using consumption data from the district, the customers that use over 100,000 cubic
gallons per year were queried out and joined to district's parcel layer. It was then spatially joined
to the pump zone layer. The pump stations that have a larger number of high consumption users
were given a higher ranking. Otay’s Water Conservation Manager reports that there were only 5
to 10 customers that the district considers high consumption users. To generate a larger sample of
users, this study identifies customers that exceed the 100,000 cubic gallons as high consumption
users (Granger, 2012). To identify these users a query was created by OWD's Database
Administrator to generate a spreadsheet of all users exceeding 100,000 cubic gallons by pump
zone. The scoring for number of high consumption users scoring range is located in Table 6.
31
Table 6 – Scoring range for number
of high consumption users.
Data Range Score
1 - 9 1
9 - 18 2
18 - 27 3
27 - 36 4
36 - 45 5
45 - 54 6
54 - 63 7
63 - 72 8
72 - 81 9
81 – 90 10
3.5.5 - Presence of a Hospital
Pressure zones were measured as to whether or not they contain a hospital. If a pressure zone
contains a hospital it would be considered relatively critical for consequences of failure. The data
for this layer were downloaded from SanGIS and spatially joined to the pressure zone layer to
determine which pressure zone contained a hospital. It was either scored as 0 for no hospital or 10
for hospital.
Table 7 – Scoring range for hospitals.
Data Range Score
0 0
1 10
3.5.6 - Serves a Major Municipality
Pressure zones are measured if any portion of the zone includes a major municipality such as the
city of Chula Vista or alternatively if zone only includes only unincorporated San Diego County.
Major municipalities are an indication of criticality because they typically contain more
customers and infrastructure than unincorporated areas. Pumps were scored either a 1 when
serving only an unincorporated area and 10 when serving any portion of a major municipality.
32
Table 8 – Scoring ranges for
municipalities.
Data Range Score
0 0
1 10
3.5.7 - Multiple Fire Stations
The uninterrupted flow of water at fire stations is crucial for pumping water into the trucks prior
to going out to fight fires. The fire station data layer is downloaded from the SanGIS website.
The fire station layer is spatially joined to the pump zone layer. The results are a trinomial
variable. If a pump station has more than one fire station within its pump zone then it is scored as
10. If a pump station has one fire station or less then the pump station is scored as a 5. If the
pump station has no fire stations it is scored as a 0.
Table 9 – Scoring ranges for fire
stations.
Data Range Score
0 0
1 5
2 10
3.6 - Non-spatial Parameters for Criticality
The non-spatial parameters of criticality include: pump redundancy lost, pump station
redundancy, reservoir redundancy, and time to restore service. The non-spatial parameters and
their scoring range are displayed in Table 10. Because these parameters are non-spatial their
scores had to be input manually into the pump's layer.
Table 10 – Scoring ranges for non-spatial features.
Criticality Factor Pump Redundancy
Lost
Pump Station
Redundancy
Reservoir
Redundancy
Time to
restore service
Ranking 2 - 10 0 - 10 0 - 10 5 - 10
33
3.6.1 - Pump Redundancy Lost
Pump redundancy is the amount of power needed by the other pumps to compensate if one of the
pumps fails. Not all pumps are working at the same time. For instance, if there are six pumps in a
pump station, usually only five will be working and the other pump is a backup in case one of the
other pumps fails. The spare pump is routinely rotated in and out with the other pumps. There is
not a specific spare pump. For instance, if there are three pumps in a pump station and pump three
is being used as the spare that does not mean it is always non operational in normal conditions. It
is rotated in and out and when it is working either pump one or pump two would be used as the
spare pump. (Stalker, 2012). In a pump station that requires the full capacity of all pumps to
operate in normal conditions, there is no back up if one of the pumps goes down. Thus pump
stations that have fewer pumps working at full capacity are considered more critical than pump
stations that have more pumps than needed to fulfill normal demands.
The scoring for redundancy lost divides the number of spare pumps and by the number of
working pumps. The formula for this process is spare pumps / functioning pumps = pump
redundancy lost. For example, in the case of pump station 944-1 it has four pumps, three working
and one spare (i.e., 1/3 = 0.33). The scoring range for Pump Redundancy Lost is found in Table
11.
Table 11 – Scoring for Pump
Redundancy Lost
Data Range Score
.20 2
.25 4
.33 6
.50 8
1 10
34
3.6.2 - Pump Station Redundancy
Pump Station Redundancy scores how many pump stations are affected when a pump fails in
another pump station. This is scored by each pump station that is served directly by another pump
station. For each pump station that is served, two points are added to the score. In Figure 9, the
944-1 pump station located in the Regulatory System shows that the 944-1 pump station in red is
the water source for 4 other pump stations. Since each pump station is scored a 2, the score for
the 944 would be 10. This process is repeated for all pumps in the district. Table 12 displays the
scoring range for Pump Station Redundancy.
Table 12 – Scoring for Pump Station
Redundancy.
Data Range Score
0 0
1 2.5
2 5
3 7.5
4 10
3.6.3 - Reservoir Redundancy
Reservoir Redundancy scores how many reservoirs are affected when a pump failure occurs.
Pumps that pump into reservoirs are less critical than those that pump to pump stations. Unlike
pump stations, reservoirs have a tank to retain water and can still distribute water even when the
pump station that pumps water to it has failed. Pumps that pump to more reservoirs than others
receive a low score because if a pump were to fail a greater volume of water has been retained as
opposed to a pump that pumped to fewer reservoirs. Table 13 displays the scoring range for
Reservoir Redundancy.
35
Table 13 – Scoring for Reservoir
Redundancy
Data Range Score
0 10
1 6.66
2 3.33
3 0
Figure 9 - Partial view of the Potable Hydraulic Profile Schematic.
Source: Otay Water District (2011)
36
3.6.4 - Time to restore service
The time to restore service is the time that it will take the district to restore service to the pump.
The type of pump is the main factor because some pumps take longer to replace than others. The
two types of pumps found in the pump stations are Vertical Turbine and Centrifugal. Centrifugal
pumps usually take longer to replace than vertical turbine pumps. Thus, for criticality centrifugal
pump will be scored as a 10 and vertical turbine pumps will be scored as a 5. The scoring range
for Time to Restore Service is located in Table 14.
Table 14 – Scoring for time to restore
service.
Data Range Score
Vertical Turbine Pump 5
Centrifugal Pump 10
The scoring range for criticality was determined by totaling all the scores for spatial and
non-spatial parameters. The data ranges were classified using equal interval classification.
Table 15 - Scoring range for
criticality.
Data Range Score
12 - 19 1
19 - 25 2
25 - 32 3
32 - 38 4
38 - 45 5
45 - 52 6
52 - 58 7
58 - 65 8
65 - 71 9
71 - 78 10
3.7 - Probability Scoring
The probability score measures how likely an asset is to fail. It is measured in terms of the age of
the asset. Using age as probability is a simple proxy for many aspects of mechanical condition.
Unfortunately, more detailed assessments of pump condition could not be included in scoring for
37
probability, because mechanical engineers did not assess the pumps during data collection. The
estimate of condition based on age is determined by subtracting the pump's age from its life
expectancy. The life expectancy for all pumps is assumed to be 15 years on average. The
following formula calculates probability for each pump: (year installed – present year) / 15.
Higher scores indicate a pump nearing the end of life expectancy with a greater likelihood to fail.
One spatial parameter that might have been considered for probability is the pressure
zone of a pump. One might hypothesize that higher pressure zones make pumps more likely to
fail than lower pressure zones. However, the pressure zone was not scored because pumps and
other infrastructure are each built to withstand the pressure it will incur no matter what zone it is
in.
Table 16 – Scoring range for
probability
Data Range Score
.06 - .41 1
.41 - .76 2
.76 – 1.10 3
1.10 - 1.45 4
1.45 - 1.80 5
1.80 – 2.14 6
2.14 – 2.49 7
2.49 – 2.84 8
2.84 – 3.18 9
3.18 – 3.54 10
3.8 - Asset Risk Determination Using ModelBuilder
Using ModelBuilder, a spatial model was created to automate the GIS process used to determine
the spatial factors for each pump. The spatial factors are determined for each pump using ArcGIS
ModelBuilder. According to Allen (2011), ModelBuilder has evolved from a simple tool into one
with many functions. The use of ModelBuilder in this case study demonstrates some of these
advancements in the tool. (For a model builder diagram developed for this study, see Appendix
38
F.) ModelBuilder was particularly useful in this study, due to its ability to automate multiple
geoprocessing tasks. To determine the risk factor, 70 geoprocessing tasks were run. Doing each
geoprocessing task individually would have been tedious and time consuming. With the
ModelBuilder all 70 tasks for this study were completed within minutes. ModelBuilder was also
used in this study as a decision support system. The non-spatial parameters were input into the
pump's layer prior to running the model. ModelBuilder then calculated the non-spatial parameters
and spatial parameters to determine criticality. Once criticality and probability were determined,
the two factors were multiplied to determine the pumps risk factor (Hyer, 2011). After the risk
factor results for the pumps were calculated, the results were analyzed to determine the accuracy
of the scores. For the overall risk factor the possible scores range from a minimum of 1 to a
maximum of 100.
39
CHAPTER 4 – RESULTS
The model developed for this thesis evaluates each of the functioning 79 potable pumps in the
OWD in terms of asset risk. The potable pumps are housed within the district's 20 pump stations.
Each station contains from 2 to 6 pumps. This information is to be used to determine priorities for
maintenance and need of repair. In Figure 10, the location of each of the 20 pump stations is
shown by a histogram in which each bar indicates the risk level of individual pumps at each
individual station. Each pump falls within one of five scoring categories after scores were
calculated by the model. In addition to displaying the pump stations, the map also displays OWD
boundary and pump zones. This map was printed out and presented to Water Systems
Supervisor.
The Water System Systems Supervisor said that the redundancy for the district is very
good, but there are a few areas that would need to be addressed immediately if pump failure were
to occur. Some of the district's pump stations are considered more pertinent than others because
they directly pump water to other pump stations. If the pump stations that provide water to others
were to fail then the pump stations that rely on other pump stations for water would be greatly
affected. Those areas are where the 944-1 pump station provides water to four other pump
stations. The other pump station is the Low Head which provides water to all of Otay Mesa. If
these two pump stations experienced some type of shutdown or failure, thousands of people could
be without water (Vaclavek, 2012).
The spatial scores were calculated according to the formula presented in Chapter 3.
Complete results are provided in Appendix A.
40
Figure 10 – Results for risk factor using equal interval rankings. Each of the 20 pump stations are
represented by a histogram and each bar represents a pump in the pump station. Source:
Alexander Schultz (2012)
Table 17 – Scoring for Risk Factor for equal
interval classification.
Data Range Score
1 - 15 LOW
15 - 29 LOW TO MODERATE
29 - 43 MODERATE
43 - 57 MODERATE TO HIGH
57 - 72 HIGH
41
4.1 - Detailed Results
To illustrate the results, the scores for five pumps from five different pump stations that received
different risk factor scores in categories ranging from high to low are displayed below in Figure
11.
Figure 11 – Chart displaying results for pumps that scored in each risk factor category.
Source: Alexander Schultz (2012)
The results for pump 1 of the Low Head pump station are displayed below in Table 18.
Probability had a slightly higher influence on risk factor than criticality. The Low Head pump
station contained the pumps with the highest risk factor scores. The area that the pump serves is
Otay Mesa which is comprised of industrial users and does not contain any single family
residential housing. The primary type of customers being served are business parks and
correctional facilities.
On the criticality indicator, this pump scored high because it serves a municipality, lacks
redundancy and has high consumption users. Since the pump was centrifugal, the pump also
received a high score for time to restore service. The area served intersects the City of San Diego
0
10
20
30
40
50
60
70
80
Low Head
980-1
711-1
1090-1
944-1
Criticality
Probablity
Risk Factor
42
boundary and received the maximum score for this parameter. The redundancy lost score was
high because the Low Head pump station only has three pumps and two backup pumps. The
pump also scored high for high consumption users because of the industrial users and the prison.
The low scoring parameters were number of schools, presence of a hospital, customers
served by pump station, pump station redundancy and reservoir redundancy. Since the land use
for this area is mainly commercial and industrial there are no schools or hospitals in this area.
The pumps for this station do not directly serve any reservoirs and only pump water to the 870-1
pump station.
The probability for this pump scored high because this pump is just past its expected
lifetime. The Water Systems Supervisor said that this pump station is to be demolished within the
next 5 years (Vaclavek, 2012).
Table 18 – Risk factor results for Pump 1 of the Low Head pumps.
Description Score
Restore Service 10
Schools Served by Pump Station 0
Presence of a Hospital 0
Customers Served by Pump Station 1
Hydrants Served by Pump Station 8
Multiple Fire Stations 5
Number of High Consumption Users 7
Serves a Major Municipality 10
Redundancy Lost 8
Pump Station Redundancy 2.5
Reservoir Redundancy 10
Criticality 61.5
Probability 3.06
Criticality after Scaling 8
Probability after Scaling 9
Risk Factor 72
The scores for pump 1 of the 980-1 pump station are displayed in table 19. The pumps for
the 980-1 pump station scored in the high to moderate category. After criticality and probability
43
were both scaled, criticality had a stronger influence on the risk factor score than probability. The
pumps for the 980-1 pump station are located in the city of Chula Vista. The customers served by
the 980-1 pump zone are single family residential.
One parameter that scored high for criticality is redundancy because the 980-1 has three
pump stations and received a redundancy score of 8, Other high scoring items for criticality
include number of schools, number of hydrants, customers served by pump station, number of
high consumption users, serves a major municipality, and reservoir redundancy.
The parameters that score low for criticality were presence of a hospital, number of
meters, number of hydrants, has multiple fire stations, number of high consumption users, is in a
municipality, pump station redundancy and reservoir redundancy. There are no hospitals within
the 980-1 pump zone. The pumps for pump station 980-1 do not directly provide water to any
pump stations.
Table 19 – Risk factor results for 980-1 pumps.
Description Score
Restore Service 5
Number of Schools 10
Presence of a Hospital 0
Customers Served by Pump Station 8
Number of Hydrants 10
Multiple Fire Stations 5
Number of High Consumption Users 10
Serves a Major Municipality 10
Redundancy Lost 8
Pump Station Redundancy 0
Reservoir Redundancy 10
Criticality 76
Probability 1.6
Criticality after Scaling 10
Probability after Scaling 5
Risk Factor 50
44
The results for pump 1 of the 711 pump station are displayed in Table 20. The pumps for
the 711 pump station scored in the moderate category. After the scores for criticality and
probability are both scaled, relatively high numbers on criticality drive the overall moderate to
high score for the 711 pump station pumps. The area served by the 711 pumps is in central Chula
Vista. The types of customers served by the pumps for the 711 were mainly single family
residential.
The criticality parameters that scored high for these pumps were number of schools,
presence of a hospital, customers served by pump station, number of hydrants, number of high
consumers, service to a municipality and reservoir redundancy. The pump zone that the 711
pumps serve has a high concentration of customers so it scores highest for customers served by
pump station, schools, and hydrants. A hospital is also located with the pump zone that the 711
pumps serve. The pump zone is also located in the City of Chula Vista, so it scored 10 for
presence of a municipality. Also there are many high consumption users because of business
parks and condominium complexes in the pump zone.
The pumps scored low for reservoir redundancy, because the 711 pumps directly serve
many reservoirs. The other parameters that scored lower were time to restore service, multiple
fire stations, redundancy lost and pump station redundancy. The pumps that make up the 711 are
vertical turbine so they do not take as long to replace. Pump station redundancy lost scores low
because the 711 has five pumps. The 711 pumps only directly serve one pump station so they
only received a score of 3.
The probability for the 711 pumps is just over their expected lifetime, but still scored
moderate compared to other pumps.
45
Table 20 – Risk factor results for Pump 1 of the 711-1 pumps.
Description Score
Restore Service 5
Number of Schools 8
Presence of a Hospital 10
Customers Served by Pump Station 10
Number of Hydrants 10
Multiple Fire Stations 5
Number of High Consumption Users 7
Serves a Major Municipality 10
Redundancy Lost 4
Pump Station Redundancy 2.5
Reservoir Redundancy 0
Criticality 71.5
Probability 1.33
Criticality after Scaling 10
Probability after Scaling 4
Risk Factor 40
The scores for pump 1 of the 1090 pump station are displayed in Table 21. The pumps for
this pump station scored in the low to moderate category. Probability of failure is a major driver
of the score due to the older age of the pumps. This pump station is located in the unincorporated
part of San Diego County and the customers served are mainly single family home residential.
The parameters that scored high for criticality were time to restore service and
redundancy lost. The pumps for the 1090-1 are Centrifugal and received a score of 10 for time to
restore service. Since there are only two pumps in the 1090-1 pump station it scored a 10 for
redundancy.
The parameters that scored low were number of schools, presence of a hospital,
customers served by pump station, number of hydrants, has multiple fire stations, number of high
consumers, is in a municipality and pump station redundancy. The pressure zone for that the
pumps for the 1090-1 serve is small and does not have any schools, hospitals, high consumption
users or fire stations. Also there are few hydrants and customers served due to the small size of
46
the pressure zone. The pump station redundancy score for 1090-1 was 0 because the pumps do
not serve water to any pump stations. The pumps in the 1090-1 are old and scored high for
probability of failure.
Table 21 – Risk factor results for 1090-1 pumps.
Description Score
Restore Service 10
Number of Schools 0
Presence of a Hospital 0
Customers Served by Pump Station 1
Number of Hydrants 1
Multiple Fire Stations 0
Number of High Consumption Users 1
Serves a Major Municipality 0
Redundancy Lost 10
Pump Station Redundancy 0
Reservoir Redundancy 3.3
Criticality 26.3
Probability 3.3
Criticality after Scaling 2
Probability after Scaling 10
Risk Factor 20
The scores for pump 1 of the 944-1 are displayed below in Table 22. The pumps for the
944-1 pump station scored in the low category. After both criticality and probability were scaled,
neither parameter had influence over the other. The pumps for the 944-1 pump station are located
in unincorporated part of San Diego County. The pump zone for that the 944-1 serves is
comprised of single family residential.
The parameters that scored high for criticality were pump station redundancy and pump
redundancy lost. The pumps for the 944-1 pump stations serves many pump stations. The pumps
received the highest score for pump station redundancy. The pump station for the 944-1 has four
pumps and received a score of 6 for redundancy lost.
47
Table 22 – Risk factor results for 944-1 pumps.
Description Score
Restore Service 5
Number of Schools 0
Presence of a Hospital 0
Customers Served by Pump Station 1
Number of Hydrants 0
Multiple Fire Stations 0
Number of High Consumption Users 0
Serves a Major Municipality 0
Redundancy Lost 6
Pump Station Redundancy 10
Reservoir Redundancy 0
Criticality 22
Probability 1.53
Criticality after Scaling 1
Probability after Scaling 5
Risk Factor 5
The parameters that scored low for criticality were number of schools, presence of a
hospital, customers served by pump station, number of hydrants, number of fire stations, and is in
a municipality. The pressure zone that the pumps for the 944-1 serve is very small and does not
serve any schools, hospitals, hydrants, fire stations. The pressure zone for these pumps is located
in unincorporated part of San Diego County. The pumps for the 944-1 are vertical turbine so the
score for time to restore service is a 5.
4.2 - Expert Confirmation of Model Results
Figure 10 was printed and then presented to Water System Supervisor. Expert confirmation
determined if the results made sense and if any of the pumps would need immediate attention
based on results. After analyzing the map and discussing the results it was determined that the
map looked good representing pumps with a few exceptions (Vaclavek, 2012). The exceptions
were largely driven by the position of the pump in the district’s network. Some of the district's
pump stations need to be considered more pertinent than others because they directly pump water
48
to other pump stations. If the pump stations that provide water to others were to fail then the
pump stations that rely on other pump stations for water would be greatly affected.
The model identified pumps that need to be replaced in the near future. This is valuable
to the district, because it will be prepared to allocate funds for future replacement and repairs.
Also if two pumps from different pump stations are due to be repaired at a certain date and the
district only has a limited amount of funds available, the district can decide which pump is more
critical to be repaired or replaced.
The scores that were found to inadequately represent the pumps position in the overall
network were those in the 944-1, Cottonwood, and the Rancho Jamul pump stations. The 944-1
had a low risk factor score, but according to Water Systems Supervisor should have scored much
higher. It should have scored higher because the pumps for the 944-1 pump station directly
provide water to four other pump stations. If the pumps in this pump station were to fail it would
leave the other pump stations without water, possibly leaving thousands of people without water.
The pumps for the Cottonwood pump station scored higher than it should have. The
station does not serve that many people. If pump failure were to occur the consequence of failure
would not be that great due to the low population that the Cottonwood pumps serve. The high
scoring result is that the pumps are old. The Water Systems Supervisor said that the pumps had
been replaced in 2011 so the model would have to be updated with the new data. Since this was
the case regarding the pumps for the Cottonwood pump station the score will be lower after
running the model again.
The pumps for the Rancho Jamul pump station scored lower than the Water Systems
Supervisor expected. The Water Systems Supervisor expected a higher score because the pumps
are starting to age and will probably be replaced within the next 10 years. In this model, the low
criticality score made the overall score lower.
49
4.3 - Sensitivity Analysis
A sensitivity analysis was conducted to better understand the model and scoring system following
feedback in the expert confirmation process. To try and increase the accuracy of the pump's risk
factor score, the classification of the data was changed from equal interval to natural breaks
(Jenks). This method is designed to set values into natural classes. The natural breaks (Jenks)
method minimizes the average deviation from class means, while maximizing the deviation of the
means from other groups.
After the model was rerun using natural breaks (Jenks) classification, the results had
changed for some of the pumps. The pumps for the 944-1 had no change and still scored in the
low risk factor. The score is still not scoring in the high risk factor category where it needs to be
given its position in the pump network. The only change is in the criticality that moves from 1 to
2 after scaling. The raw criticality score (i.e., before scaling) remained the same but because
natural breaks (Jenks) groups the scores differently when scaling the scores.
The scoring for the 1090-1 pumps increased from low scoring to low-moderate scoring
using natural breaks (Jenks). Another sensitivity analysis is to weight a few parameters to see if
scores can be improved. The scores for pumps in the 944-1 pump station are the least accurate.
The reason for this is that the pump zone that the pumps provide water for is small, so most of the
spatial parameter scores were very low or 0. The scores that the pumps for the 944-1 scored high
in was pump station redundancy and reservoir redundancy. Since these are the only 2 parameters
for which the 944-1 pumps had high scores, they will be increased to help increase the risk factor
score and improve the results. Pump station redundancy and reservoir redundancy will be
increased by multiplying each score by 10. The goal of this will be to increase the risk factor
score for the pumps of the 944-1 pump station but not increase or decrease the other pump risk
50
factor scores. In this weighting process the natural breaks (Jenks) scaling is retained to classify
the results, since it solved one of the problems noted earlier with the score.
Figure 12 – Results for risk factor using natural breaks (Jenks) classification. Source: Alexander
Schultz (2012)
51
Table 23 – Scoring range for Risk Factor using
natural breaks (Jenks).
Data Range Score
4 - 10 LOW
10 - 18 LOW TO MODERATE
18 - 30 MODERATE
30 - 50 MODERATE TO HIGH
50 - 90 HIGH
Table 24 – Risk factor results for 944-1 pumps equal interval and natural breaks (Jenks).
Description Equal Interval Natural Breaks (Jenks)
Restore Service 5 5
Number of Schools 0 0
Has a Hospital 0 0
Customers Served by Pump Station 1 1
Number of Hydrants 0 0
Multiple Fire Stations 0 0
Number of High Consumption Users 0 0
Is In a Municipality 0 0
Redundancy Lost 6 6
Pump Station Redundancy 10 10
Reservoir Redundancy 0 0
Criticality 22 22
Probability 1.53 1.53
Criticality after Scaling 1 2
Probability after Scaling 5 5
Risk Factor 5 10
Increasing the pump station redundancy and the reservoir redundancy by multiplying by
10 had both a positive and a negative impact on the risk factor scores. The positive impacts were
the pumps for the 711-1 returned to moderate to high risk factor. This is good because the 711-1
pumps are seen as critical by OWD staff because of the high population that is served. The
negative impacts were that the risk factor score for the 944-1 did not change; they still remained
at moderate to high risk factor. This was not good because the 944-1 pumps are seen as critical by
OWD staff because of the pump stations they directly provide water to. Also the risk factor scores
52
for the pumps of the Cottonwood pump station increased from moderate to moderate to high
scores.
Figure 13 – Results for Risk Factor using natural breaks (Jenks) and weighting redundancy x 10.
53
Table 25 – Scoring for risk factor range for
redundancy x 10.
Data Range Score
8 - 16 LOW
16 - 32 LOW TO MODERATE
32 - 50 MODERATE
50 - 80 MODERATE TO HIGH
80 - 100 HIGH
Table 26 – Risk factor results for 944-1 pumps increasing x 10.
Description Score
Restore Service 5
Number of Schools 0
Has a Hospital 0
Customers Served by Pump Station 1
Number of Hydrants 0
Multiple Fire Stations 0
Number of High Consumption Users 0
Is In a Municipality 0
Redundancy Lost 6
Pump Station Redundancy 100
Reservoir Redundancy 100
Criticality 212
Probability 1.53
Criticality after Scaling 10
Probability after Scaling 5
Risk Factor 50
54
Figure 14 – Risk factor scoring results after populating 944-1 pumps’ spatial parameters with
mean score.
Table 27 – Scoring for risk factor for pumps
with mean criticality.
Data Range Score
4 - 9 LOW
9 - 18 LOW TO MODERATE
18 - 30 MODERATE
30 - 50 MODERATE TO HIGH
50 - 90 HIGH
55
Table 28 – Risk factor results for 944-1 pumps with mean criticality.
Description Score
Restore Service 5
Number of Schools 3
Has a Hospital 1
Customers Served by Pump Station 5
Number of Hydrants 6
Multiple Fire Stations 3
Number of High Consumption Users 5
Is In a Municipality 3
Redundancy Lost 6
Pump Station Redundancy 3
Reservoir Redundancy 5
Criticality 45
Probability 1.53
Criticality after Scaling 5
Probability after Scaling 5
Risk Factor 25
To increase the risk factor score for the 944-1 pumps, the mean scores for all spatial
parameters were then added to the spatial parameters of the 944-1 pumps. Populating the 944-1
pumps' spatial scores with the mean from the other pumps still did not increase the risk factor
score to place into a higher scoring category. The overall criticality score increased but not
enough to move it out of a moderate risk factor.
Since the mean did not help increase the overall risk factor score the next step will be to
populate each of the 944-1 pumps spatial parameters with the maximum possible score. This
should increase the overall criticality for the 944-1 pumps to the maximum possible score. No
matter what was done to increase the 944-1 pumps scoring, the results were still not high enough
to put it in the highest scoring range.
56
Figure 15 – Risk factor scoring results after populating 944-1 pumps’ spatial parameters with
maximum score.
Table 29 – Scoring range for risk factor with
maximum criticality.
Data Range Score
4 - 9 LOW
9 - 18 LOW TO MODERATE
18 – 30 MODERATE
30 - 50 MODERATE TO HIGH
50 - 90 HIGH
57
Table 30 – Risk factor results for 944-1 pumps with maximum criticality.
Description Score
Restore Service 10
Number of Schools 10
Has a Hospital 10
Customers Served by Pump Station 10
Number of Hydrants 10
Multiple Fire Stations 10
Number of High Consumption Users 10
Is In a Municipality 10
Redundancy Lost 10
Pump Station Redundancy 10
Reservoir Redundancy 10
Criticality 110
Probability 1.53
Criticality after Scaling 10
Probability after Scaling 5
Risk Factor 50
Figure 16 – Chart displaying different scoring results for 944-1 pumps Source: Alexander Schultz
(2012)
0
5
10
15
20
25
30
35
40
45
50
Equal
Interval
Natural
Breaks
Natural
Breaksx10
Natural
Breaks
mean
Natural
Breaks max
Criticality
Probabilty
Risk Factor
58
CHAPTER 5 – CONCLUSIONS
Implementing an asset management system using GIS can be a very powerful tool for managing
and predicting risk factors for both vertical and horizontal assets. As was discussed previously,
asset management can be very beneficial and cost efficient for water utilities. Further work is
required to refine the model presented in this study. However, the model generated in this study
brings some immediate benefits to the OWD and also demonstrates the value of integrating GIS
with asset management for other utilities.
5.1 - Model Refinements
One important parameter used for determining probability that was not included was condition.
Probability only had one parameter of age. After trying three different strategies to boost the
overall risk factor score of the 944-1 pumps, none of them was able to increase the risk factor
score enough to place it into the high risk factor category. Possibly if a condition assessment was
included it could have increased the pump's probability score and raised its own risk factor score
or possibly lowered it depending on the condition.
Other possible strategies would be increasing the number of parameters used to
calculate the model results. Since the pump zone for the 944-1 pumps is not very large, non-
spatial parameters would most likely be used. These non-spatial parameters could include the size
of the reservoir for which the pumps provide water and the 944-1 distributes water to other pump
stations. The more pumps a pump station has, the larger amount of water it uses to distribute to
other customers. Each pump that directly provides water to another pump station could inherit a
weighted proportion of all the spatial and non-spatial criticality and probability risk factors
scoring from the pumps for which it provides water. The weighted proportions could be based on
the allocation of the water provided to other pumps vs. directly to customers. This would be
similar to feeder streams and tributaries flowing into a major river with a network diagram that
59
would resemble a trunk and branches of a tree. The greater the number and size of the tributaries
the greater flow of water is discharged into the river, increasing the rivers size. The metaphor is
instructive because river systems are quite complex and dynamic as a tributaries discharge is not
always constant. For example they may experience seasonal fluctuations (Snavely, 2006).
Such complexity can be found in the OWD network of pump stations because demand for
water across the network is not constant. Water demand can also change during different times of
the day and during different seasons. A dynamic algorithm might be required to calculate
parameters for allocation. Also having an application that can calculate and monitor this
procedure would also assist in determining this parameter’s output. Since I did not have the data
for peak and off peak demand or the application to calculate and monitor it, I did not use this
parameter. It also involved the complexity of calculating the demand needed by 944-1 pumps to
supply the other pumps and how much water each reservoir can store and distribute if pump
failure were to occur.
5.2 - Benefits to Otay Water District
One of the benefits that this model brings for the OWD in the way of proactive maintenance is
that the district can plan and prepare for when an asset is projected to fail. Also the district will be
prepared for the cost of repairing or replacing the asset when life expectancy is coming to an end.
Proactive maintenance can also be planned to prolong an asset’s life expectancy and save the
district future replacement costs. The new information that this brings to the Water Operations
department is pumps that might not have been seen as critical, may now have to be inspected for
condition assessment to determine if pumps are possibly close to failure. An example of this is
the pumps for the 870-1 pump station were inspected and found to be aging and in need of
replacement. The CIP that was going to be used for future replacement of the pump station was
scheduled to be in 5 years. After the pump station was found to be aging and outdated, the Water
60
Systems Supervisor is trying to have the Capital Improvement Project (CIP) moved up to 3 years
instead of 5. Another example is the pumps for the 980-1 pump station that scored as Moderate to
High. After being inspected, it was found that the pumps were starting to show age and would
need to be replaced fairly soon. Since the 980-1 pump station is strictly now being used as a
backup for the 980-2, replacing these pumps is currently not urgent.
The new insights that the model brings to the district is a better understanding of spatial
influences on a given pump’s criticality. The consequence of failure is defined as the impact on
service to customers. The model creates a better understanding of how some pumps take
precedence above others in the way of repairs or replacement and suggests which pumps are
priorities to investigate for mechanical condition and to perform repairs to prolong life
expectancy.
The study highlights the importance of spatial aspects for asset management of vertical
assets. The environment surrounding the asset has an important influence on the overall asset risk
score. Infrastructure asset data are typically identified, associated with, or referenced by their
geographic locations and spatial relationships. As a result, GIS and spatial data analysis can play
to support asset management processes (Halfawy and Figueroa, 2006).
The spatial location of a pump can not only influence its risk factor score but also other
pumps’ risk factor scores. The 944-1 pump’s risk factor score is influenced by the parameters in
its pump zone. Since there are very few meters, hydrants, and high consumption parcels, the
criticality score for the 944-1 pumps is very low.
5.3 - Otay Accomplishments Present and Future
What OWD has accomplished so far in its implementation of asset management is creating a GIS
geodatabase that can store not only horizontal assets but also vertical assets. Vertical assets and
their attributes that in the past were not present in GIS can now be accessed. Also the asset
61
management data entry form was created so operations crew members who are out in the field
taking asset inventory can populate the database in the field. Currently at this time data are still
being acquired and a condition assessment for each vertical asset still has yet to be completed.
OWD is also in the process of implementing Azteca’s City Works for its CMMS. The
main reason for using City Works is its geo-centric platform. Unlike other CMMS products, its
platform sits on top of the GIS database and does not require a separate database for data storage.
Having a separate GIS and CMMS to constantly keep in sync with each other can be frustrating
and tedious because each database is being populated by different sources. With this process of
keeping both databases in sync eliminated, the district can now save time and money. City
Works will be used to keep track of maintenance, create work orders and maintain an inventory
of assets.
The case study that was conducted in the previous chapters to find the risk factor of assets
required a model to be created in ArcGIS ModelBuilder. This process was successful in finding
the risk factor for pumps, but is not an ideal long term solution for determining risk factor for all
assets. Different models would have to be created for each asset, because not all assets would
have the same parameters for determining risk factor. Each model would need to be maintained in
sync between GIS and the CMMS. To solve this problem, OWD will implement a new
application that will automate the risk factor scoring system. This application will integrate with
GIS and City Works. It will predict when an asset will fail and determine what the repair or
replacement costs will be. This application will calculate each asset’s risk factor with its own
customized interface which will let the user add parameters they want to use to determine an
asset’s risk factor. Also the application will send out alerts when an asset in nearing the end of its
lifecycle or when it will need future maintenance. Once in the system the asset will constantly be
monitored during its lifecycle, unlike a model in ModelBuilder where the model would constantly
62
be rerun in order to generate new results. OWDs ultimate goal is to have a fully automated asset
management system.
5.4 - Value of Integrating GIS with Asset Management in Utilities
Other utilities that provide electricity or gas may also benefit from the database and scoring
model that includes not only horizontal assets but also vertical assets, like the one described here.
Although some are reported in literature as using GIS to find risk factor for only their horizontal
assets, very few discuss using a geodatabase to store vertical feature class tables. Since power
stations like pump stations are only being represented by a point in GIS, the vertical assets are
excluded. For example, there are a large number of pipelines in power stations, most of which are
underground or overhead and constitute complex networks. Using GIS, users can collect the
information on the geographical distribution of pipelines or overhead transmission systems and
corresponding service areas (Shahidehpour, 2005). When speaking about GIS in an energy utility
very few discuss storing a vertical asset table in the geodatabase model. This would be very
beneficial because then utilities would also be able the access their vertical asset data within GIS,
which could then be edited or used in geoprocessing functions to help determine risk factor.
Other energy utilities do not use GIS to track their assets. Progress Energy’s Lee Plant in
Goldsboro, NC could have benefited by using GIS model to tracks their vertical pump’s asset risk
factor. Instead they use reactive rather than a proactive approach. As the pumps aged their
maintenance became more frequent and more expensive. Several years ago their reliability had
deteriorated to the point of compromising the overall plant operations (Anonymous, 2007). Had
they had an asset management system in place the pumps could have been monitored and
proactive maintenance or replacement could have been done to reduce costs.
63
REFERENCES
Allen, D. 2011. Getting to Know ArcGIS: ModelBuilder. Redlands, CA: Esri Press.
American Society of Civil Engineers (ASCE). 2009. ASCE Report Card for America's
Infrastructure.
http://www.infrastructurereportcard.org. (Accessed: June 24, 2012).
American Water Works Association (AWWA). 2012. Buried no Longer: Confronting America's
Water Infrastructure Challenge.
http://awwa.org/infrastructure. (Accessed June 27, 2012).
Anonymous. 2007. Focus on O&M: Asset Management. Power
http://search.proquest.com.libproxy.usc.edu/docview/232492849 (Accessed: July 1, 2012).
Australia and New Zealand National Asset Management Steering Group (NAMS). 2011.
International Infrastructure Management Manual. New Zealand: NAMS Group.
AWI. 2010. Asset management becoming the water wave of the future. American Water
Intelligence. 1:3
http://www.americanwaterintel.com/archive/1/12/market-insight/asset-management-becoming
-water-wave-of-future.html (Accessed June 24, 2012).
Baird, G. 2011. Building a Solid Foundation During an Enterprise Asset Management
Revolution. Water Utility Infrastructure Management.
http://www.uimonline.com/index/webapp-stories-action=511 (Accessed: Feb. 21, 2012).
Crothers, H. 2010. GIS and Enterprise Asset Management.
http://blogs.Esri.com/dev/blogs/waterutilities/archive/2010/03/04 (Accessed: June 27, 2012).
Crothers, H blog posted 3/4/2011. How GIS Supports the Core Business Patterns of a
Water Utility.
http://blogs.Esri.com/dev/blogs/waterutilities/archive/2011/03/04 (Accessed: June 27, 2012).
Doyle, M and Rose, D. 2001. Protecting your Assets. Water Environment and Technology.
13:7.
Esri. 2011. Water Utilities Data Model.
http.support.Esri.com/en/downloads/datamodel/detail/16 (Accessed: June 23, 2012).
Granger, W. 2012. Interview by author. Spring Valley CA. March 10.
64
Halfawy, M and Figueroa, R. 2006. Developing Enterprise GIS-Based Data Repositories for
Municipal Infrastructure Asset Management. Joint International Conference on Computing and
Decision Making in Civil and Building Engineering. Montreal, June 14, 2006.
http://www.nrc-cnrc.gc.ca.libproxy.usc.edu/obj/irc/doc/pubs/nrcc45583/nrcc45583.pdf
(Accessed: July 5, 2012).
Harlow, K. 2000. Part 1: Asset management: A key competitive strategy.
News for Public Works Professionals.
http://www.publicworks.com/content/news/article.asp?DocID={187E5CA1-8FC4-11D4-8C6
(Accessed: Feb. 21, 2012).
Harlow, K. 2000. Part 4: Asset management: Getting Serious. News for Public Works
Professionals.
http://www.publicworks.com/content/news/article.asp?DocID={734C4132-E1B9-11D4-A76F
(Accessed: Feb. 21, 2012).
Hyer, C. 2010. Improving Utility O&M and Capital Decisions by Incorporating the Concepts
of Asset Condition, Criticality and Risk. WEFTEC 2010 Proceedings, Session 1-10, pp 167-175.
http: usc.illiad.oclc.org/iliad/pdf/409994.pdf. (Accessed: Feb. 21, 2012).
Hyer, C. 2011. Building a Comprehensive Asset Management Program using Consistent
Condition, Consequence of Failure and Risk Scoring. Texas Water Convention, 5-8 April,
Fort Worth, TX.
http://www. tawwa.org/TW11Paper/celinehyer.pdf. (Accessed: Feb. 21, 2012).
Lutchman, R. 2006. Sustainable Asset Management: Linking Assets, People, and
Processes for Results. Lancaster, PA: DEStech Publications, inc.
McKibben, J and Davis, D. 2002. Integration of GIS with Computerized Maintenance
Management Systems (CMMS) and Asset Management Systems. 22nd Annual ESRI
International User Conference, 8-12 July, San Diego, CA.
http://proceedings.Esri.com/library/userdconf/proc02/pap0554/p0554.htm.
(Accessed: June 24, 2012).
New Mexico Environmental Finance Center. 2006. Asset Management: A Guide for Water
and Wastewater Systems.
http://www.nmenv.state.nm/us/dwb/assistance/documents/AssetManagementGuide.pdf.
(Accessed: June 27, 2012).
Otay Water District. 2008. Otay Water District Strategic Plan: Fiscal Years 2009-2011.
http://www.otaywater.gov/extranet/apps_stratplan3.0/ppt/measures/stratplan.pdf.
(Accessed: June 27, 2012).
Otay Water District. 2012. About Otay.
http://www.otaywater.gov/Otay/page.aspx?g=33 (Accessed: June, 30, 2012)
65
Shahidehpour, M and Ferrero, R. 2005. Time Management for Assets:
Chronological Strategies for Power System Asset Management. Power and Energy Magazine,
IEEE
http://ieeexplore.ieee.org.libproxy.usc.edu/xpls/abs_all.jsp?arnumber=1436498. (Accessed: June
27, 2012)
Shamsi, U. 2005. GIS Applications for Water, Wastewater and Stormwater Systems. New
York: Taylor and Frances.
Shamsi, U. 2002. GIS Tools for Water, Wastewater and Stormwater Systems. New York:
Taylor and Francis.
Sinha, S. n.d. Municipal Water and Wastewater Infrastructure Asset Management Primer.
Penn State Department of Civil and Environment Engineering.
https://courses.worldcampus.psu.edu/public/buried_assets/files/primer.pdf.
(Accessed: June 24, 2012).
Snavely, K. 2004. Rivers. The Gale Encyclopedia of Science. Ed. K. Lee Lerner and Brenda Wilmoth
Lerner. 3rd ed. Vol. 5. Detroit: Gale, 2004. 3460-3464. Gale Virtual Reference Library.
http://go.galegroup.com.libproxy.usc.edu/ps/i.do?action=interpret&id=GALE%7CCX341850196
8&v=2.1&u=usocal_main&it=r&p=GVRL&sw=w&authCount=1
(Accessed: July 1, 2012).
Stalker, G. 2012. Interview by author. Spring Valley, CA. June 13.
U.S. General Accounting Office (GAO). 2004. Water infrastructure: Comprehensive Asset
Management Has Potential to Help Utilities Better Identify Needs and Plan for Future
Investments (GAO-04-461).
http://www.gao.gov/products/GAO-04-461. (Accessed: Feb. 21, 2012).
U.S. Environmental Protection Agency (EPA). April, 2008a. Asset Management: A Best
Practices Guide (EPA 816-F-08-014).
http://www.epa.gov/ogwdooo/smallsystems/pdf/guide_smallsystems_assetmanagement_
bestpractices.pdf
(Accessed: Feb. 21, 2012).
U.S. Environmental Protection Agency (EPA). June, 2008b. Effective Utility Management: A
Primer for Water and Wastewater Utilities.
http://epa.gov/scitech/wastetech/upload/tool_sc_waterum_primer for effectiveutilities.pdf.
(Accessed: Feb. 21, 2012).
Vaclavek, J. 2012. Interview by author. Spring Valley, CA. March 2.
66
Vanier, D. 2004. Geographic Information Systems (GIS) as an Integrated Decision
Support Tool for Municipal Infrastructure Asset management. National Research Council
Canada.
http://www.nrc-cnrc.gc.ca/obj/ircdoc/pubs/nrcc4675/nrcc754.pdf. (Accessed: Feb. 21, 2012).
Water Infrastructure Network (WIN). 2001. Water Infrastructure NOW.
http://win-water.org/reports/windows.pdf. (Accessed: June 27, 2012).
Zhao, M and Stevens, G. 2002. Implementing GIS for a Water Distribution at Otay
Water District. International Users Conference, ESRI. San Diego, CA. July, 2002
http://proceedings.Esri.com/library/userconf/proc02/pop0572.htm. (Accessed: June 24, 2012).
Zhao, M and Stevens, G. 2003. Elevating GIS to an Enterprise Solution at Otay Water district.
International Users Conference, ESRI, San Diego, CA. July, 2003
http://proceedings.esri.com/libratry/userconf/proc03/p0213.pdf. (Accessed: June 24, 2012).
Zhao, M and Stevens, G. 2009. ROI through GIS: Cutting Cost and Time. Geospatial Today.
28:4.
Zhao, M and Stevens, G. 2011. GIS and Asset Management: A Cost Effective
Enterprise Solution. 2011 Annual ESRI International Users Conference, San Diego, CA. July, 2003.
http://proceedings.esri.com/library/userconf/proc11/papers/2088_200.pdf. (Accessed June 24, 2012).
67
APPENDIX A: Table displaying results using equal interval classification.
Description
Type
Time to restore service
Redundancy Lost
Pump Station Redundancy
Reservoir Redundancy
Schools
Hospitals
Customers Served
Hydrants Served
Fire Stations
High consumption users
Is in a municipality
Criticality
Probability
Risk Factor
944-1-1
Vertical
turbine
5 6 10 0 0 0 1 0 0 0 0 1 5 50
944-1-2
Vertical
turbine
5 6 10 0 0 0 1 0 0 0 0 1 5 50
944-1-3
Vertical
turbine
5 6 10 0 0 0 1 0 0 0 0 1 5 50
944-1-4
Vertical
turbine
5 6 10 0 0 0 1 0 0 0 0 1 5 50
1296-1-1
Vertical
turbine
5 4 7.5 0 4 0 1 2 10 2 0 3 4 12
1296-1-2
Vertical
turbine
5 4 7.5 0 4 0 1 2 10 2 0 3 4 12
1296-1-3
Vertical
turbine
5 4 7.5 0 4 0 1 2 10 2 0 3 4 12
1296-1-4
Vertical
turbine
5 4 7.5 0 4 0 1 2 10 2 0 3 4 12
1296-1-5
Vertical
turbine
5 4 7.5 0 4 0 1 2 10 2 0 3 4 12
978-1-1
Vertical
turbine
5 6 0 3.3 0 0 1 2 5 2 0 1 10 10
978-1-2
Vertical
turbine
5 6 0 3.3 0 0 1 2 5 2 0 1 4 4
978-1-3
Vertical
turbine
5 6 0 3.3 0 0 1 2 5 2 0 1 10 10
978-1-4
Vertical
turbine
5 6 0 3.3 0 0 1 2 5 2 0 1 4 4
980-1-1
Vertical
turbine
5 8 0 10 10 0 8 10 5 10 10 10 5 50
980-1-2
Vertical
turbine
5 8 0 10 10 0 8 10 5 10 10 10 5 50
980-1-3
Vertical
turbine
5 8 0 10 10 0 8 10 5 10 10 10 5 50
711-1-1
Vertical
turbine
5 4 2.5 0 8 10 10 10 5 7 10 10 4 40
711-1-2
Vertical
turbine
5 4 2.5 0 8 10 10 10 5 7 10 10 4 40
711-1-3
Vertical
turbine
5 4 2.5 0 8 10 10 10 5 7 10 10 4 40
711-1-4
Vertical
turbine
5 4 2.5 0 8 10 10 10 5 7 10 10 4 40
711-1-5
Vertical
turbine
5 4 2.5 0 8 10 10 10 5 7 10 10 4 40
Rancho Jamul-1
Vertical
turbine
5 6 0 6.6 0 0 1 1 0 0 0 1 7 7
Rancho Jamul-2
Vertical
turbine
5 6 0 6.6 0 0 1 1 0 0 0 1 7 7
Rancho Jamul-3
Vertical
turbine
5 6 0 6.6 0 0 1 1 0 0 0 1 7 7
68
Rancho Jamul-4
Vertical
turbine
5 6 0 6.6 0 0 1 1 0 0 0 1 7 7
850-2-1
Vertical
turbine
5 4 2.5 0 2 0 4 5 10 4 0 3 3 9
850-2-2
Vertical
turbine
5 4 2.5 0 2 0 4 5 10 4 0 3 3 9
850-2-3
Vertical
turbine
5 4 2.5 0 2 0 4 5 10 4 0 3 3 9
850-2-4
Vertical
turbine
5 4 2.5 0 2 0 4 5 10 4 0 3 3 9
850-2-5
Vertical
turbine
5 4 2.5 0 2 0 4 5 10 4 0 3 3 9
803-1-1
Vertical
turbine
5 2 7.5 0 4 0 3 4 5 3 0 3 2 6
803-1-2
Vertical
turbine
5 2 7.5 0 4 0 3 4 5 3 0 3 2 6
803-1-3
Vertical
turbine
5 2 7.5 0 4 0 3 4 5 3 0 3 3 9
803-1-4
Vertical
turbine
5 2 7.5 0 4 0 3 4 5 3 0 3 3 9
803-1-5
Vertical
turbine
5 2 7.5 0 4 0 3 4 5 3 0 3 3 9
803-1-6
Vertical
turbine
5 2 7.5 0 4 0 3 4 5 3 0 3 3 9
832-1-1
Vertical
turbine
5 8 5 3.3 1 0 1 1 0 1 0 2 4 8
832-1-2
Vertical
turbine
5 8 5 3.3 1 0 1 1 0 1 0 2 4 8
832-1-3
Vertical
turbine
5 8 5 3.3 1 0 1 1 0 1 0 2 4 8
832-1-4
Vertical
turbine
5 8 5 3.3 1 0 1 1 0 1 0 2 4 8
832-1-5
Vertical
turbine
5 8 5 3.3 1 0 1 1 0 1 0 2 4 8
832-1-6
Vertical
turbine
5 8 5 3.3 1 0 1 1 0 1 0 2 4 8
Pointe-1
Vertical
turbine
5 8 0 10 0 0 1 1 0 0 0 1 2 2
Pointe-2
Vertical
turbine
5 8 0 10 0 0 1 1 0 0 0 1 2 2
Pointe-3
Vertical
turbine
5 8 0 10 0 0 1 1 0 0 0 1 2 2
Low Head-1 Centrifugal 10 8 2.5 10 0 0 1 8 5 7 10 8 9 72
Low Head-2 Centrifugal 10 8 2.5 10 0 0 1 8 5 7 10 8 9 72
Low Head-3 Centrifugal 10 8 2.5 10 0 0 1 8 5 7 10 8 9 72
Rolling Hills-1
Vertical
turbine
5 6 0 10 0 0 1 1 0 0 10 3 1 3
Rolling Hills-2
Vertical
turbine
5 6 0 10 0 0 1 1 0 0 10 3 1 3
Rolling Hills-3
Vertical
turbine
5 6 0 10 0 0 1 1 0 0 10 3 1 3
Rolling Hills-4
Vertical
turbine
5 6 0 10 0 0 1 1 0 0 10 3 1 3
980-2-1
Vertical
turbine
5 2 0 0 10 0 8 10 5 10 10 8 2 16
980-2-2
Vertical
turbine
5 2 0 0 10 0 8 10 5 10 10 8 2 16
69
980-2-3
Vertical
turbine
5 2 0 0 10 0 8 10 5 10 10 8 2 16
980-2-4
Vertical
turbine
5 2 0 0 10 0 8 10 5 10 10 8 2 16
980-2-5
Vertical
turbine
5 2 0 0 10 0 8 10 5 10 10 8 2 16
980-2-6
Vertical
turbine
5 2 0 0 10 0 8 10 5 10 10 8 2 16
870-1-1
Vertical
turbine
5 4 7.5 10 0 0 1 8 5 7 10 7 10 70
870-1-2
Vertical
turbine
5 4 7.5 10 0 0 1 8 5 7 10 7 10 70
870-1-3
Vertical
turbine
5 4 7.5 10 0 0 1 8 5 7 10 7 10 70
870-1-4
Vertical
turbine
5 4 7.5 10 0 0 1 8 5 7 10 7 10 70
870-1-5
Vertical
turbine
5 4 7.5 10 0 0 1 8 5 7 10 7 10 70
Cottonwood-2 Centrifugal 10 10 0 10 4 0 3 4 5 3 0 6 6 36
Cottonwood-1 Centrifugal 10 10 0 10 4 0 3 4 5 3 0 6 6 36
1090-1-1 Centrifugal 10 10 0 3.3 0 0 1 1 0 1 0 2 10 20
1090-1-2 Centrifugal 10 10 0 3.3 0 0 1 1 0 1 0 2 10 20
1530-1-1 Centrifugal 10 8 0 10 0 0 1 1 0 0 0 2 4 8
1530-1-2 Centrifugal 10 8 0 10 0 0 1 1 0 0 0 2 4 8
1530-1-3 Centrifugal 10 8 0 10 0 0 1 1 0 0 0 2 4 8
1200-1-1
Vertical
turbine
5 8 0 6.6 0 0 1 1 0 1 0 1 7 7
1200-1-2
Vertical
turbine
5 8 0 6.6 0 0 1 1 0 1 0 1 7 7
1200-1-3
Vertical
turbine
5 8 0 6.6 0 0 1 1 0 1 0 1 7 7
1004-1-1
Vertical
turbine
5 8 2.5 10 0 0 1 1 0 0 0 2 2 4
1004-1-2
Vertical
turbine
5 8 2.5 10 0 0 1 1 0 0 0 2 2 4
1004-1-3
Vertical
turbine
5 8 2.5 10 0 0 1 1 0 0 0 2 2 4
1485-1-1
Vertical
turbine
5 8 0 3.3 1 0 1 1 0 1 0 1 1 1
1485-1-2
Vertical
turbine
5 8 0 3.3 1 0 1 1 0 1 0 1 1 1
1485-1-3
Vertical
turbine
5 8 0 3.3 1 0 1 1 0 1 0 1 1 1
70
APPENDIX B: Integration of the spreadsheet into data model.
71
APPENDIX C: Asset Data Entry Form showing examples of parent assets.
72
APPENDIX D: Example of Asset Data Entry Form for pump station.
73
APPENDIX E: Feature class relating to abstract class.
74
APPENDIX F: Diagram of model used to determine risk factor.
Abstract (if available)
Abstract
This study demonstrates the integration of Geographic Information Systems (GIS) with asset management. There are few existing studies or demonstrations of the integration of GIS technology with asset management systems, especially for vertical assets at water utilities. A model is developed using Otay Water District (OWD) as a case study. The case study expands upon a GIS model that already contains horizontal assets (e.g., pipelines). The new model includes vertical assets (e.g., pump stations). In the past, non-spatial vertical assets, such as pump stations and their components were represented only by a point and could not be plotted against spatial data variables. In the expanded model, spatial and non-spatial asset risk variables are measured and scored for the 79 pumps within the 20 pump stations at the district. Each pump is assigned criticality and probability scores, which are then multiplied to give an overall risk factor score. Model scores were plotted on a point symbology map and expert confirmation was conducted with OWD water operations staff. A sensitivity analysis of the model reveals that manipulating model parameters to increase overall scoring accuracy of some pumps can also have a negative impact on the scoring of others. Further study is needed to plan and implement schemes that allow vertical assets at utilities to inherit asset management scores based on their positions within larger horizontal networks.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
A Web GIS application for airport pavement management
PDF
The role of GIS in asset management: County of Kauai Department of Parks and Recreation a need for an asset management program
PDF
Integrating land survey data into measurement-based GIS: an assessment of challenges and practical solutions for surveyors in Texas
PDF
Using GIS and asset management to understand hydrant damages and required maintenance
PDF
A comparison of GLM, GAM, and GWR modeling of fish distribution and abundance in Lake Ontario
PDF
Spatiotemporal patterns of salt and nutrient contamination in Los Angeles County's groundwater basins
PDF
Evaluating transit and driving disaggregated commutes through GTFS in ArcGIS
PDF
Geospatial web application development to access irrigation asset data: Veterans Affairs Palo Alto Health Care System
PDF
Integrating spatial visualization to improve public health understanding and communication
PDF
A model for placement of modular pump storage hydroelectricity systems
PDF
Implementing spatial thinking with Web GIS in the non-profit sector: a case study of ArcGIS Online in the Pacific Symphony
PDF
Assessing the impact of a web-based GIS application to promote earthquake preparation on the University of Southern California University Park Campus
PDF
Modeling nitrate contamination of groundwater in Mountain Home, Idaho using the DRASTIC method
PDF
Modeling prehistoric paths in Bronze Age Northeast England
PDF
Address points and a maser address file: improving efficiency in the city of Chino
PDF
Remote analysis of avalanche terrain features: identifying routes, avoiding hazards
PDF
The geography of voter power in the U.S. electoral college from 1900-2012
PDF
Applying GIS to landscape irrigation systems: a case study of the Music Academy of the West campus in Montecito, CA
PDF
Utilizing GIS and remote sensing to determine sheep grazing patterns for best practices in land management protocols
PDF
Using pattern oriented modeling to design and validate spatial models: a case study in agent-based modeling
Asset Metadata
Creator
Schultz, Alexander John
(author)
Core Title
The role of GIS in asset management: integration at the Otay Water District
School
College of Letters, Arts and Sciences
Degree
Master of Science
Degree Program
Geographic Information Science and Technology
Publication Date
09/11/2012
Defense Date
09/10/2012
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
asset management,GIS,OAI-PMH Harvest
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Kemp, Karen K. (
committee chair
), Vos, Robert (
committee chair
), Wilson, John (
committee chair
)
Creator Email
ajschult@usc.edu,pbsurfer2002@yahoo.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-96520
Unique identifier
UC11289085
Identifier
usctheses-c3-96520 (legacy record id)
Legacy Identifier
etd-SchultzAle-1190.pdf
Dmrecord
96520
Document Type
Thesis
Rights
Schultz, Alexander John
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
asset management
GIS