USF Libraries
USF Digital Collections

Development of a GIS based infrastructure replacement prioritization system

MISSING IMAGE

Material Information

Title:
Development of a GIS based infrastructure replacement prioritization system a case study
Physical Description:
Book
Language:
English
Creator:
Pickard, Brian D
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Asset
Utility
IMS
IMMS
Main
Rehabilitation
Hydraulic
Model
Dissertations, Academic -- Environmental Engineering -- Masters -- USF
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Maintenance, repair, and replacement of transmission mains and distribution system piping is expected to cost approximately $75 billion over the next two decades to ensure that public water systems are capable of providing the United States with safe drinking water. However, there is a significant gap between the funds available and the projected costs of infrastructure replacement or rehabilitation. Infrastructure Management Systems (IMS) have been developed to assist utilities and decision-makers in determining how to allocate resources for infrastructure. This project utilizes theTampa Water Department (TWD) as a case study to develop a tool for prioritizing infrastructure replacement.TWD is responsible for managing over 2,240 miles of pipeline. Building booms in the 1920s and 1950s have inadvertently resulted in a significant need to replace or rehabilitate pipelines due to the aging of the overall water supply infrastructure. To address this problem, TWD is taking th e first steps in applying IMS to transmission anddistribution pipelines. Currently, approximately 500 miles of water mains have been slated for replacement or rehabilitation. The TWD has a GIS that has been used to map and integrate information on main breaks, service line breaks, customer complaints and modeled water age. Information on fire hydrant spacing and line flushing dates are also integrated into the GIS. Following development of the GIS based infrastructure replacement prioritization system, approximately 3,000 pipe segments were identified and queries were performed to help develop cost to benefit analyses. The results were used to develop a prioritized list of potential capital projects and incorporate the time value of money and event forecasting. The GIS was also used to develop indicators of the overall infrastructure condition. From this analysis it was possible to develop an approach to categorize projects and identify the resources needed to address high priority pro blems associated with undersized mains, unlined cast iron mains, asbestos cement mains, and hydraulic looping projects. As water infrastructure rehabilitation and replacement needs increase in the future, the need for adaptable methods to prioritize capital spending will also increase.This study has demonstrated the ability to prioritize long-term and short-term infrastructure projects using a GIS platform in conjunction with databases and spreadsheets.
Thesis:
Thesis (M.A.)--University of South Florida, 2006.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Brian D. Pickard.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 68 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001790316
oclc - 144608070
usfldc doi - E14-SFE0001496
usfldc handle - e14.1496
System ID:
SFS0025815:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001790316
003 fts
005 20070618091119.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 070618s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001496
040
FHM
c FHM
035
(OCoLC)144608070
049
FHMM
090
TA170 (ONLINE)
1 100
Pickard, Brian D.
0 245
Development of a GIS based infrastructure replacement prioritization system :
b a case study
h [electronic resource] /
by Brian D. Pickard.
260
[Tampa, Fla] :
University of South Florida,
2006.
3 520
ABSTRACT: Maintenance, repair, and replacement of transmission mains and distribution system piping is expected to cost approximately $75 billion over the next two decades to ensure that public water systems are capable of providing the United States with safe drinking water. However, there is a significant gap between the funds available and the projected costs of infrastructure replacement or rehabilitation. Infrastructure Management Systems (IMS) have been developed to assist utilities and decision-makers in determining how to allocate resources for infrastructure. This project utilizes theTampa Water Department (TWD) as a case study to develop a tool for prioritizing infrastructure replacement.TWD is responsible for managing over 2,240 miles of pipeline. Building booms in the 1920s and 1950s have inadvertently resulted in a significant need to replace or rehabilitate pipelines due to the aging of the overall water supply infrastructure. To address this problem, TWD is taking th e first steps in applying IMS to transmission anddistribution pipelines. Currently, approximately 500 miles of water mains have been slated for replacement or rehabilitation. The TWD has a GIS that has been used to map and integrate information on main breaks, service line breaks, customer complaints and modeled water age. Information on fire hydrant spacing and line flushing dates are also integrated into the GIS. Following development of the GIS based infrastructure replacement prioritization system, approximately 3,000 pipe segments were identified and queries were performed to help develop cost to benefit analyses. The results were used to develop a prioritized list of potential capital projects and incorporate the time value of money and event forecasting. The GIS was also used to develop indicators of the overall infrastructure condition. From this analysis it was possible to develop an approach to categorize projects and identify the resources needed to address high priority pro blems associated with undersized mains, unlined cast iron mains, asbestos cement mains, and hydraulic looping projects. As water infrastructure rehabilitation and replacement needs increase in the future, the need for adaptable methods to prioritize capital spending will also increase.This study has demonstrated the ability to prioritize long-term and short-term infrastructure projects using a GIS platform in conjunction with databases and spreadsheets.
502
Thesis (M.A.)--University of South Florida, 2006.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 68 pages.
590
Adviser: Robert P. Carnahan, Ph.D.
653
Asset.
Utility.
IMS.
IMMS.
Main.
Rehabilitation.
Hydraulic.
Model.
690
Dissertations, Academic
z USF
x Environmental Engineering
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1496



PAGE 1

Development of A GIS Based Infrastructure Replacement Prioritization System; A Case Study by Brian D. Pickard A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Environmental Engineering Department of Civil and Environmental Engineering College of Engineering University of South Florida Major Professor: Robert P. Carnahan, Ph.D. Audrey D. Levine, Ph.D. Paul Zandbergen, Ph.D. Date of Approval: March 30, 2006 Keywords: asset, utility, IMS, IMMS, main, rehabilitation, hydraulic, model Copyright 2006, Brian D. Pickard

PAGE 2

DEDICATION This project is dedicated to thos e before me who have built a strong foundation for this project, have willfully supported me during this project regardless of personal sacrifices, and to those having the drive to improve upon this project for the greater good.

PAGE 3

ACKNOWLEDGEMENTS The Tampa Water Department (TWD) has been 100% supportive of this project by providing both tangible and intang ible resources. I am particularly thankful to several TWD employees have been particularly benefici al to this project including: Harlan Reynolds and Rory Jones who initially introduced me to the potential benefits of utilizing GIS to prioritize water infrastructure improvements. Harlan Reynolds has been particularly helpful in sharing AutoCAD techniques saving me hours of labor and preventing avoidable mistakes. Jeff Hough has been extremely helpful by demonstrating methods to integrate MapInfo with AutoCAD as well as demonstrating how queries are performed in the GIS environment. The Graduate Committee for this project has provided valuable insight and recommendations without making this project unnecessarily burdensome. Dr. Levine in particular has devoted many hours towards project direction, organization and formatting. Dr. Carnahan and Dr. Levine ha ve also greatly assisted in providing administrative guidance in obtaining fi nal approval for this project.

PAGE 4

Special Note to the Reader The original document contains color that is necessary for unde rstanding the data. The original thesis is on file with the USF library in Tampa, Florida.

PAGE 5

i TABLE OF CONTENTS LIST OF TABLES.............................................................................................................iv LIST OF FIGURES.............................................................................................................v NOMENCLATURE........................................................................................................viii ABSTRACT....................................................................................................................... ..x INTRODUCTION...............................................................................................................1 OBJECTIVES..................................................................................................................... .7 LITERATURE REVIEW....................................................................................................8 Infrastructure needs..................................................................................................8 Accounting challenges.............................................................................................9 Infrastructure failure costs.....................................................................................10 Water infrastructure replacement...........................................................................11 Asset management.................................................................................................12 Geographic information systems (GIS).................................................................13 Infrastructure condition databases.........................................................................14 Integrated infrastructure management systems......................................................15 METHODOLOGY............................................................................................................17 Phase 1 Policy decisions.....................................................................................18 Master plan programs................................................................................18

PAGE 6

ii Prioritization elements...............................................................................19 Survey........................................................................................................21 Phase 2 Data collectio n, mapping and formatting..............................................22 Data collection...........................................................................................22 Construction costs......................................................................................24 Phase 3 Structured queries and database development.......................................24 Structured queries......................................................................................25 Database development...............................................................................26 Phase 4 Project prioritization..............................................................................28 RESULTS........................................................................................................................ ..31 Phase 1 Policy decisions.....................................................................................29 Phase 2 Data collectio n, mapping and formatting..............................................30 Data collection...........................................................................................32 Construction costs......................................................................................42 Phase 3 Data extraction.......................................................................................42 Structured queries......................................................................................44 Database development...............................................................................45 Phase 4 Project prioriti zation and master planning.............................................45 DISCUSSION....................................................................................................................4 9 Prioritization system..............................................................................................49 Integrated database.................................................................................................50 Database applications.............................................................................................51

PAGE 7

iii Condition indicators...............................................................................................51 Resource management...........................................................................................52 CONCLUSION..................................................................................................................54 ENGINEERING IMPLICATIONS...................................................................................56 Procedural changes................................................................................................56 Improved resources to support decisions...............................................................59 ADDITIONAL RESEARCH.............................................................................................60 REFERENCES..................................................................................................................64 BIBLIOGRAPHY..............................................................................................................67

PAGE 8

iv LIST OF TABLES Table 1: Comparison of four master plan programs that were developed to address particular issues related to a certain type of pipe.............................................19 Table 2: Summary of data for the TW D service area and the time frame of data availability........................................................................................................20 Table 3: Matrix linking master plan programs and prioritization elements as determined by survey of TWD policy makers.................................................21 Table 4: Description of a ttribute data variable types us ed for database manipulation..23 Table 5: Overlay of benef it units of each prio ritization element with master plan combinations to allow for calcula tion of benefit to cost ratios........................28 Table 6: Results of Tampa Water Department survey to determine benefit factor values ..............................................................................................................31 Table 7: Cost factors utilized for various nominal pipe diameters ranging from 4-36 inches...............................................................................................................42 Table 8: Dimensions of buffer regions fo r each type of prioritization element.............44 Table 9: Relative annual changes in water ma in breaks, service line breaks, pressure complaints and water quality complaints.........................................................47 Table 10: Benefit unit based costs util ized in the prioritization process..........................47

PAGE 9

v LIST OF FIGURES Figure 1: The Tampa Water Department se rvice area covers a 211 square mile area......1 Figure 2: The illustrated planned infrastruc ture project is contained within a buffer region. The buffer region associates planned projects with the location of prioritization elements ......................................................................................4 Figure 3: Area of water di stribution system with a propos ed hydraulic looping project. The buffer region dimensions are increased to encompass a hydrant for system flushing..............................................................................................................5 Figure 4: Area of water di stribution system with an unlined water main causing red water complaints. The buffer region di mensions are increased to encompass the complaint locations that result ed from the unlined water main...................6 Figure 5: Design overview of research project................................................................17 Figure 6: Example of SQL script utilized to associate prioritization elements with planned projects...............................................................................................26 Figure 7: Example of the count command in MS Access to combine databases. Note the query is written to in clude all planned projects in the results table...........27 Figure 8: Total Length of Rehabilitation & Replacement Projects by Type (miles)......33 Figure 9: Relative location of various type s of planned rehabilitation, replacement and improvement projects within the TWD service area.......................................34

PAGE 10

vi Figure 10: Water main break locations within the TWD service area..............................35 Figure 11: Service main break locati ons within the TWD service area............................36 Figure 12: Water main flushing locati ons within the TWD service area..........................37 Figure 13: Water quality complaint locat ions within the TWD service area....................38 Figure 14: Water pressure complaint lo cations within the TWD service area..................39 Figure 15: Proposed fire hydrant locatio ns within the TWD service area........................40 Figure 16: Results of water age mode ling within the TWD service area..........................41 Figure 17: Estimated construction costs for planned TWD rehabilitation, replacement and improvement projects................................................................................43 Figure 18: Visual representation of planned project buffer regions and prioritization elements...............................................................................45 Figure 19: Comparison of average quantities of prioritization element occurrences amon planned rehabilitation, replacement and improvement projects......................46 Figure 20: Results yield benefit to cost ratios ranging from 0.35 to 0.00. A relatively small portion of projects have benefit to cost ratios greater than 0.10............48 Figure 21: Comparison of existing rehabilitatio n and replacement funding levels and the ideal funding levels as determined by the top 10% of prioritized projects determined by this project................................................................................53 Figure 22: Procedures are needed to conti nually maintain geospatial databases when customer driven projects and capital projects occur simultaneously. The process flow diagram above shows input from several functional groups are needed to accomplish this task.........................................................................58

PAGE 11

vii Figure 23: Example of inaccurate assignments of prioritization elements due to the use of buffer regions........................................................................................61

PAGE 12

viii NOMENCLATURE Benefit factor: A policy generated dimensionless va lue expressing a relative importance to a water utility. Benefit unit: A cost associated with a future occu rrence of a prioritization element. This value is determined by utilizing tangible cost s in conjunction with policy decisions. CMMS : Computerized Maintenance Management System GIS : Geographic Information System IMS : Infrastructure Management System IIMS : Integrated Infrastruc ture Management System LRS : Linear Referencing System Master plan program : A group of projects expected to accomplish a specific goal having value to a water utility.

PAGE 13

ix Prioritization element: A category of spatial data that can have either tangible or no tangible value to a water utility. R&R : Rehabilitation and replacement TWD : Tampa Water Department

PAGE 14

x Development of A GIS Based Infrastructure Replacement Prioritization System; A Case Study Brian D. Pickard ABSTRACT Maintenance, repair, and replacement of transmission mains and distribution system piping is expected to cost approximate ly $75 billion over the next two decades to ensure that public water systems are capabl e of providing the United States with safe drinking water. However, there is a signifi cant gap between the f unds available and the projected costs of infrastructure repla cement or rehabilitati on. Infrastructure Management Systems (IMS) have been devel oped to assist utilitie s and decision-makers in determining how to allocate resources for infrastructure. This project utilizes the Tampa Water Department (TWD) as a case st udy to develop a tool for prioritizing infrastructure replacement. TWD is responsible for managing over 2,240 miles of pipeline. Building booms in the 1920s and 1950s have inadvertently result ed in a significant need to replace or rehabilitate pipelines due to the aging of the overall water supply infrastructure. To address this problem, TWD is taking the firs t steps in applying IMS to transmission and distribution pipelines. Curre ntly, approximately 500 miles of water mains have been slated for replacement or rehabilitation. The TWD has a GIS that has been used to map and integrate information on main breaks, service line breaks, customer complaints and

PAGE 15

xi modeled water age. Information on fire hydr ant spacing and line fl ushing dates are also integrated into the GIS. Following development of the GIS based in frastructure replacement prioritization system, approximately 3,000 pipe segments were identified and queries were performed to help develop cost to benefit analyses. Th e results were used to develop a prioritized list of potential capital projects and inco rporate the time value of money and event forecasting. The GIS was also used to deve lop indicators of the overall infrastructure condition. From this analysis it was possibl e to develop an appr oach to categorize projects and identify the resources needed to address high priority problems associated with undersized mains, unlined cast iron ma ins, asbestos cement mains, and hydraulic looping projects. As water infrastructure rehabilitation and replacement needs increase in the future, the need for adaptable methods to prio ritize capital spending wi ll also increase. This study has demonstrated the ability to prioritize long-term and short-term infrastructure projects using a GIS plat form in conjunction with databases and spreadsheets.

PAGE 16

1 INTRODUCTION The Tampa Water Department (TWD) admi nisters the City of Tampas water utility. The service area covers an area 211 square miles in size and serves approximately 510,000 residents. The TWD service ar ea is shown in Figure 1. Figure 1. The Tampa Water Department service area covers a 211 square mile area.

PAGE 17

2 The City of Tampa water supply system consists of supply, treatment, storage, pumping, transmission, distribution and service. Treated water is distributed to fire mains, 12,543 fire hydrants and 124,371 meters. The transmission system consists of 268 miles of pipe having a nominal diameter of 16 inches to 54 inch es. The distribution system consists of more than 2,007 miles of pi pe ranging in size from 2-inch to 14-inch. Pipes are made of cast iron (CI), ductile iron (DI), asbestos cement (AC), high-density polyethylene (HDPE), and polyvinyl chloride (PVC). The Tampa Water Department recognizes the need for planning infrastructure rehabilitation, replacement and improvement. Master plans identif y vulnerable portions of the service area where water system probl ems are most likely to occur and elucidate preventive measures to maintain service le vels. These plans are used extensively for budgeting and planning purposes. This thesis is focused on four TWD master plans. 1. Undersized Main Replacem ent (UMR) Master Plan 2. Unlined Cast Iron Main Replacement (UCIMR) Master Plan 3. Hydraulic Looping System Master Plan 4. Asbestos Cement Main Replacem ent (ACMR) Master Plan Approximately 2,900 infrastructure related projects are identified by the master plans with an approximate construction co st of $390 million in 2006 dollars. However, only approximately $6 million is available on an annual basis. Thus, a mechanism is

PAGE 18

3 needed to prioritize rehab ilitation, replacement and improvement projects based on a rational basis. One tool that can be applied to projec t prioritization is Geographic Information Systems (GIS) software. Information on infras tructure condition indicators such as main breaks, service main breaks, customer comp laints and water age are collected by the TWD. However, these data are not compiled in a way that allows for a comprehensive view of historical trends in infrastructure rehabilitati on, replacement and improvement projects. GIS software can be used to inte grate these databases into a single location thereby allowing for prioritization with a goal of providing improved service to rate payers. The starting point is to es tablish prioritization elemen tsa category of spatial data that has tangible or no ta ngible value to a water utilit y. Buffer regions are then established surrounding planned rehabilitati on, replacement and improvement projects. A buffer region is necessary because the pr ioritization element occurrences are not entered into GIS software at exactly the same location as the planned projects. Typically, databases containing information such as main breaks, customer complaints, and flushing locations contain addresses not coordinates. These address-based databases are linked to a database containing the coordinates for all addresses within the service area. Because the assigned coordinates for each address is not linked to the water main serving the address, the size of the buffer region is esta blished so the occurrenc e of a prioritization element can be linked to a planned project. This concept is illustrated in Figure 2.

PAGE 19

4 Figure 2. The illustrated planned infrastructure project is contained within a buffer region. The buffer region associates pla nned projects with the location of prioritization elements. For some prioritization elements, it may be necessary to increase the buffer region dimensions. For example, it is important that the proposed re gion generated around a planned hydraulic looping projec t contains the flush point lo cations. However, projects intended to eliminate flushing requirements are not necessarily near hydrant locations that can be used for water main flushing. An example of increasing the buffer region dimension is shown in Figure 3.

PAGE 20

5 Figure 3. Area of water distribut ion system with a proposed hydraulic looping project. The buffer region dimensions are increased to encomp ass a hydrant for system flushing. It is also useful to increase buffer re gion dimensions when associating customer complaints with planned infrastructure project s. Pressure or water quality complaints can be caused by pipes located at different locati ons than the complaint itself. Red water, a common water quality concern related with unlin ed pipe, can originate on one street and flow into other locations. An example of how pipes can cause water quality issues in a larger area is shown in Figure 4.

PAGE 21

6 Figure 4. Area of water dist ribution system with an unlined water main causing red water complaints. The buffer region dimensions are increas ed to encompass the complaint locations that resulted from the unlined water main. The use of GIS and appropriate buffer regions allows for linking different information sources in a format that can allow for prioritization of rehabilitation, replacement and improvement projects. This integrated approach enables a utility to prioritize projects based on a hi erarchy of factors with the ultimate result of providing improved service and rate stabili zation to water customers.

PAGE 22

7 OBJECTIVES The purpose of this project is to inves tigate the feasibility of applying GIS to prioritize water infrastructure rehabilitation, replacement a nd improvement projects for the City of Tampa Water Department There are four objectives. 1. Convert existing infrastructure re placement master plans into an electronic format incorporating at tribute data to permit geospatial analysis. 2. Consolidate existing databases incl uding main break reports, service main break reports, customer complaint logs and flushing reports into a common format that can be utilized by GIS software. 3. Perform geospatial analyses to determine a benefit to cost ratio for each planned project based on a prioritization matrix agreed upon by Tampa Water Department policy makers. 4. Prepare a combined, prioritized infrastructure replacement master plan and make appropriate budget recommendations.

PAGE 23

8 LITERATURE REVIEW Infrastructure needs Most water utilities place most avai lable resources on distribution system expansions as opposed to ensuring the sustaina bility of existing systems. Inadequate funding and few support technologies within th e United States contribute to the current need to address aging water infrastructure (Vanier, Danylo and Ville de Montreal Finance Department, 1998) This need is quant ified in two reports to congress prepared by the United States Environmental Pr otection Agency (USEPA). A 1997 report prepared by the USEPA incl udes the results of a water utility survey including 4,000 community water systems. Conducted in 1995 and 1996, the survey addresses infrastructure needs for community water systems to comply with the Safe Drinking Water Act. Although th e 1997 USEPA report does not estimate replacement and rehabilitation needs, it does estimate the per centage of need related to water distribution and transmi ssion piping is 56 percent of th e total $138.4 billion 20-year need. (USEPA, 1997) A second water utility survey conducted in 1999 by the USEPA shows the total infrastructure need is likely underestimated by the 1997 Drinking Water Infrastructure Needs Survey. The 1999 survey in dicates the total existi ng need to comply

PAGE 24

9 with the Safe Drinking Water Act, excludi ng rehabilitation and re placement needs is $150.9 billion. (USEPA, 2001) Two reports have compared current sp ending levels with the estimated funding required to provide a safe drinking water supp ly. This difference in funding levels is known as the water infrastructure funding g ap. The Clean Water and Drinking Water Infrastructure Gap Analysis estimates an annual $12 billion funding deficiency exists within the United States. (USEPA, 2002) A nother report prepared by the Congressional Budget Office estimates the gap to be be tween -$0.2 billion and $8.3 billion annually depending on future revenue stream increas es. (Congressional Budget Office, 2002) Water customers currently receive an ex cellent value when compared to other utilities. Water, sewer, and solid waste charges combined cost customers less than 0.8 percent of total household expe nditures. When compared to 2.4 percent and 2.1 percent for electric and telecommunicati on expenditures respectively, the costs of water, sewer, and solid waste are small. (Beecher, 2001) Accounting challenges One cause of the funding gap is due to the method water utilities account for buried infrastructure. Alt hough water transmission and distribution systems account for approximately 80 percent of assets owned by a water utility (Grablutz and Hanneken, 2001), they are excluded from the general fund regardless if they have been purchased with general fund dollars. This is because transmission and distribu tion system piping is

PAGE 25

10 not considered to be a means to meet debt service or obtain revenue. Because this infrastructure is not considered a general fund asset, the effect of the depreciation is often overlooked when analyzing the fi nancial strength of water uti lities. (Lemer, 1998) This method does not encourage enacting methods to assure buried water infrastructure remains a sustainable asset. Infrastructure failure costs The real value of infrastructure lies, however in the services it provides, its enabling role in supporting other economic a nd social activities. (Lemer, 1998) Thus the value of water main infrastructure does not equal the replacement cost and can be both tangible and non-tangible. Efforts to prioritize infrastructure repl acements and improvements are historically based on economic approaches. Recent litera ture suggests it is a ppropriate to also include non-tangible values. The Decision S upport System and the Grand Central Model, developed by the American Water Works Association Research Foundation (AwwaRF) include operational, social, and other fact ors with non-tangible value. (Zhang, 2004) Another AwwaRF project cla ssifies costs as follows: Costs Incurred Directly by the Utility 1. Administrative and legal co sts of damage settlements 2. Lost product costs 3. Public safety costs

PAGE 26

11 4. Repair and return to service costs 5. Service outage mitigation costs 6. Utility emergency response costs Routine (high incidence-low impact) Social Costs 1. Access impairment and travel delay costs 2. Customer outage and substitution costs Low Incidence/High Impact Social Costs 1. Health damages 2. Direct damage at the point of failure 3. Waterborne illness introduced as a result of failure 4. Property damages 5. Reduced fire fighting costs The sum of all costs above is re ferred to as the Total Societal Cost. Perhaps the greatest challenge associated with incorporating non-tangible costs is assigning corresponding numerical values. This causes difficulty in generating a one si ze fits all model. (Hasan, 2002) Water infrastructure replacement Literature recommends utilities to fund maintenance and repairs at approximately 2 to 4 percent of their asset replacement value. (Vanier, Danylo, and Ville de Montreal Finance Depart ment, 1998) Approximately $1.74 b illion is spent annually in

PAGE 27

12 the United States for water infrastructure rehabilitation or replacement. Various techniques utilized include cleaning and lining, pipe lining and slip lining. (Weston, 2002) Most utilities have not yet experienced the financial effects of replacing large quantities of infrastructure. However, resear ch indicates water infr astructure is maturing, and a new era is on the horizon known as the Replacement Era. (GAO and AWWA, 2001) The public will ultimately fund these replacements if the current level of service is expected to remain unchanged. (Neukrug, 2002) Asset management Asset management is the process of keeping track of and deploying the publics capital. (Lemer, 1998) Effective asse t management programs contain tools for minimizing costly emergency repairs, ma king strategic funding decisions designed to keep rates low and bond ratings high, measur e the efficiency and effectiveness of maintenance programs, defending and protec ting cash reserves for future asset R&R expenditures and meeting new accounting and environmental regulatory standards. (Anderson and Smith, accessed 2006) One elem ent that assists in accomplishing the above goals is project priori tization. Effectively prioriti zing projects can maximize returns, stabilize rate s, decrease bond inte rest rates and enhance communication among stakeholders. (Nagel and Elenbass, 2006) However, the quantity of buried water main infrastructure makes prioritizing project s extremely problematic. (AWWA, 2004)

PAGE 28

13 AwwaRF recommends considering the ag e and material of pipe, number and frequency of main breaks, reduction in hydrau lic capacity, water quality problems, joint types and related joint leaks, strategic consid erations such as pavement overlay programs, soil conditions and pipe depth, customers th at cannot have their service interrupted, number and frequency of [fittings] and [quantity] of service connections when prioritizing the replacement of buried water main infrastructure. (Weston, 2002) Utilizing factors such as these can result in efficiency savings. The key to effective asset management is considering as many rele vant factors as possible when making maintenance and replacement choices. (CBO, 2002) Geographic information systems (GIS) Considering various factors for a larg e number of water main infrastructure projects inherently involves vast data quant ities. Managing this data is of primary importance and requires a relatively large effo rt to perform this task sufficiently. (Matichich, Allen and Allen, 2006) One critical type of data management that must be accomplished to effectively priori tize water infrastructure proj ects is associating attribute data to a geographic location. This powerful tool is becoming increasingly popular with governmental agencies. (Lemer, 1998) The Seattle Public Utilities utilizes GIS in a Sewer Pipe Risk Model to prioritize sewer infrastructure replacements (Martin, 2005) and the PIPES system to evaluate water pipelines. The PIPES syst em combines a criticality rating and a

PAGE 29

14 deterioration rating to effectively manage infr astructure rehabilitati on and replacement. (Lim and Pratti, 1997) The St. Louis Count y Water Company uses a similar method by utilizing GIS to compare a water main conditi on database with histor ical water pipeline work orders. Direct costs are calculated from the wate r main condition database and indirect costs are estimated to be approxima tely 20 to 40 percent of direct costs. (Grablutz and Hanneken, 2001) Infrastructure condition databases Although GIS is becoming increasingly popul ar, the majority of water utilities do not utilize software packages designed to faci litate decisions regard ing pipe rehabilitation or replacement. (Weston, 2002) However, there currently exists many commercialized computerized maintenance management syst ems (CMMS). These systems are extremely capable in storing data associated with infr astructure condition, but are historically not perfect with respect to water distribution syst em life cycle analysis, risk analysis, and project prioritization. (Vanier, Danylo and Ville de Montreal Finance Department, 1998) Several of these packages are tailored to a specific infrastructure type, such as water distribution system infrastructure. (Lemer 1998) The KANEW model in particular helps asset managers address the timing of infrastructure replacement on a macro-scale level. The KANEW model does not address in dividual pipe segments. (Grablutz and Hanneken, 2001) Asset management software and infrastructure condition databases are beginning to incorporate probabi listic failure, the time value of money and the balance

PAGE 30

15 between theoretically ideal replacement time s and real world budget limitations. (Nagel and Elenbass, 2006) Integrated infrastructure management systems (IMMS) Infrastructure Management Systems (IMS) an element of an asset management program, are currently utilized separately for ea ch infrastructure classe.i. water, sewer, storm water and transportation. The first IM S originated with pavement management systems (PMS) and bridge management syst ems (BMS). A switch in management paradigms to consider each infrastructure cl ass collectively as part of an Integrated Infrastructure Management System (IIMS) is on the horizon. (Ferre ira and Duarte, 2005) Lemer proposes IIMS due to inefficienci es associated with considering each infrastructure class independently. The inefficiencies are widespread and easy to see: jammed traffic on roads designed to carry only a fr action of the current demand, newlyresurfaced city streets ripped open to re pair aged subsurface pipes, news media expressing outrage th at traffic lanes must be closed for maintenance or that basements are flooded. (Lemer,1998) It is clear individual infrast ructure classes are closely integrated and the overall cost to the rate paying public can be redu ced by addressing each infrastructure class collectively. An IIMS can be designed to combine information from multiple platforms or establish a single platform reducing the difficulties in merging multiple data formats.

PAGE 31

16 Regardless of the chosen method, an IIMS should consider each infras tructure class when facilitating asset management decisi ons. (Ferreira and Duarte, 2005)

PAGE 32

17 METHODOLOGY This project is organized into four phases Phase 1 is designed to establish items of value for project prioritizat ion. The data to support th e items of value identified during Phase 1 are gathered, mapped and format ted in a geospatial format incorporating attribute data during Phase 2. During Phase 3, the geospatial data are associated with planned projects in a single database. The da tabase created during Phase 3 is utilized in conjunction with items of value identified in Phase 1 to prioritize pl anned projects during Phase 4. Detailed methods for each phase ar e included below. The interrelationships among the four phases is illustrated in Figure 5. Figure 5. Design overview of research project.

PAGE 33

18 Phase 1 Policy decisions Prioritizing infrastructure rehabilitati on or replacement projects inevitably involves policy decisions. Master plan pr ograms are created to a ddress certain issues often related to a particular type of pipe. These issues can also be called prioritization elements because the extent that each issue is present with each planned project can be used as a basis to prioritize the projects w ithin master plan programs. This project surveys water utility policy makers to dete rmine the importance of each prioritization element for each master plan program. Master plan programs The Tampa Water Department has histor ically categorized rehabilitation, replacement or improvement projects in four categories as defined in Table 1.

PAGE 34

19Table 1. Comparison of four ma ster plan programs that were develo ped to address particular issues related to a certain type of pipe. Master Plan Program Definition Undersized Main Replacement These projects consist of the replacement of water mains having a diameter of less than 6 inches. Much of this pipe is galvanized which has decreased Cfactors due to tuberculation. Water mains of these sizes are not capable of providing fire flows meeting modern standards and are considered a public health and safety concern. Unlined Cast Iron Main Replacement Iron and other metals can leach out of unlined pipe creating tuberculation. Excessive tuberculation is associated with red water and poor C-factors. Red water leads to customer complaints and low Cfactors ultimately result in inadequate fire flow. This pipe is replaced with lined ductile iron pipe under this program. Due to metal leaching this type of pipe is also susceptible to structural failures leading to water main breaks. Hydrualic Looping System Projects Without flushing, dead end water mains cause increased water ages ultimately leading to customer complaints. This type of project loops dead end mains together to facilitate water flow and decrease water age. These projects also increase available fire flow and decrease costs associated with water main flushing. Asbestos Cement Main Replacement Although not considered a public safety concern from a water supply standpoint, this pipe is extremely brittle and breaks easily. Tapping this type of pipe also requires specific procedures to prevent exposure to airborne asbestos. These projects replace asbestos cement pipe eliminating maintenance and environmental concerns. Prioritization elements Because the age of the infr astructure and the pace of development, 2,886 projects have been identified that can be cate gorized as rehabilitation, replacement or improvement projects. To prioritize these pr ojects and allocated resources, it is necessary to define specific items of value. These items of value, or prioritization elements are

PAGE 35

20 defined as a type of spatial data that can have tangible or intangible value to water utility rate payers. The prioritization elements that have a geospatial component available for this project are listed in Table 2. Table 2. Summary of data for the TWD service area and the time frame of data availability. Geospatial Component Time Frame Available for Analysis Water Main Breaks FY 2001 through FY 2005 Service Main Breaks FY 2001 through FY 2005 Approximate Water Age FY 2004 demand allocation Flushing logs FY 2004 and FY 2005 Customer Pressure Complaints FY 2001 through FY 2005 Customer Water Quality Complaints FY 2001 through FY 2005 Undersized mains are not sized to provi de fire protection. The location of additional fire hydrants necessary to meet the fire hydrant spacing requirement of the utility can be used as a prioritization el ement for evaluating undersized mains. The TWD policy is for fire hydrant spacing is f ire hydrants shall be no more than 450 feet apart when measured along streets or accepta ble access ways. For dead-end cul-de-sacs, fire hydrants shall be placed no more than 450 f eet from the rear of the farthest structure (TWD, 2002). Currently, compliance with this standard is assesse d through a tedious manual technique. Proposed new hydrant locat ions are divided into two categories:

PAGE 36

21 those that can be installed without water ma in improvements and those that will need water main improvements to achieve av ailable flows for fire protection. Survey Survey participants including the TWD di rector, deputy director, chief engineer, chief planning engineer and ch ief design engineer are aske d to rank the importance of various prioritization elements for each type of master plan program. Each participant is instructed to assign a weighted value to each of the prioritization elements in each of the master plan programs on a percentage basis. The data is compiled to yield a prioritization matrix that can be presented to utility policy makers. The survey itself takes the form of Table 3. Table 3. Matrix linking master plan programs and prioritization elements as determined by survey of TWD policy makers. Percent Importance Master Plan Program Water Main Breaks Service Main Breaks Quantity of Proposed New Hydrants Modeled Water Age Quantity of Flushing Visits Water Pressure Complaints Water Quality Complaints Hydraulic Looping Undersized Mains Unlined Cast Iron Asbestos Cement

PAGE 37

22 Phase 2 Data collection and construction costs Data describing planned projects and pr ioritization elements are collected and formatted so the various databases can be eff ectively utilized for a common purpose. The geospatial component of the projects and prio ritization elements is then determined so GIS analysis can be performed. Construction costs are estimated for each nominal pipe diameter to aid in later analysis. Data collection To develop the framework for this project, it is necessary to convert master plan programs from a hardcopy format to an electr onic format with geos patial components and attribute data. When converting the individu al pipe segments found in each master plan program to an electronic format, it is critical that an appropriate coordinate system is used. For example, the coordinate system used by TWD is the USA NAD83, Florida State Plane, West Zone, US-foot coordinate system. Attribute data must be attached to each planned project to allow for informa tion retrieval by personnel not utilizing GIS software. The attribute data attached as pa rt of this project ar e given in Table 4.

PAGE 38

23 Table 4. Description of attribute data va riable types used for database manipulation. Attribute Data Description Data Type Master plan program Type of project (undersize main, hydraulic looping, etc.) Text String Proposed nominal pipe diameter Pipe diameter listed in inches Integer "Along" Street Common street name parallel to the planned project. Text String "From" Street Street intersecting the Along Street at the beginning boundary of the planned project Text String "To" Street Street intersecting the Along Street at the ending boundary of the planned project Text String Associated atlas page TWD utilizes a numbered atlas system based on a one mile grid pattern Text String Water age Estimated average day time elapsed from the treatment facility to a specific location in the distribution system Text String or Real Number Pipe installation date (if available) Calendar year when the pipe that is to be replaced by a planned project was installed Integer In each case, Autodesk Land Desktop 2006 is used to enter data and define the attribute data. Water age is determined by running an average day extended period simulation in a hydraulic modeling software package. Th e simulation length should be the maximum water age expected in the system plus a la rge safety margin. The run time length is checked for appropriateness by comparing the water age at several simulation intervals towards the end of the simulation to assure the calculated water age values remain constant. The model links and nodes are then color coded by calcula ted water age at the final simulation time increment. Regions c ontaining calculated water age increments are then created in AutoCAD and the water age increment is then associated with each planned rehabilitation, replacement or impr ovement project as attribute data.

PAGE 39

24 Due to the relative importance of proposed new hydrants on planned undersized main projects, a fire hydrant placement master plan is created as part of this project. The proposed fire hydrant locations were divide d into two broad categories. The first category is proposed fire hydr ants that can be installed with no water main improvements. The second category is proposed fire hydrants insta llations that would require water main improvements. Proposed fire hydrant locations are determined utilizing TWD fire hydrant spacing criteria. Construction costs Estimated construction costs on a per foot basis are calculated for various pipe sizes to allow later calculations of estimated pr oject costs. An estimated cost per foot is determined for each nominal pipe diameter for both grassed and paved areas. A single average cost per linear foot can then be es timated from historical utility data on the percentages of water mains being installed under street pavement or within grassed areas. Phase 3 Structured queries and database development Once all data is collected and formatted into a geospatial databa se, GIS is utilized to extract the occurrence of each prioriti zation element for each planned project. Once prioritization element occurrences are linked to a planned project, the resulting databases are then combined into a master database suitable for the prioritization process.

PAGE 40

25 Structured queries Once prioritization elements and indivi dual projects are combined into an electronic format in a common coordinate sy stem, data can be extracted in accordance with the prioritization matrix developed in Phase 1 using a GIS software package. First, the data collected in Phase 2 are imported into Ma pInfo Professional keeping the coordinate systems consistent. Appropriate buffer region dimensions are then established and a reference key for each planned replacem ent, rehabilitation or improvement project is then added to the corresponding buffer regi on. Once the identifyi ng key is generated, it is important to not change the table of planned capital projects midstream during the prioritization process. Otherwise the databa se will not associate pr ioritization elements with the correct planne d capital project. The buffer region table with the de sired buffer length is then opened simultaneously with the prioritization element of interest. A Stru ctured Query Language (SQL) script is then written to select all occurrences of a prioritization element within each of the planned capital project buffer region s. An example of how this SQL script is written in MapInfo is shown in Figure 6.

PAGE 41

26 Figure 6. Example of SQL script utilized to associ ate prioritization elemen ts with planned projects Database development The results of this query are then exported to an Access data base for further analysis. Because there may be more than one occurr ence of a prioritizati on element within the associated buffer region for each planned capital project, it is necessary to employ the

PAGE 42

27 Count command in Access to determine the number of prioritization element occurrences for each planned cap ital project. An example of an MS Access query is shown in Figure 7. Figure 7. Example of the count command in MS Access to combine databases. Note the query is written to include all planned projects in the results table. After obtaining the number of occurrences for each planned project, the identifying key is then used to generate a single table contai ning all planned capital projects with the corresponding number of occurrences for each prioritization element. The summary table provides input for the prioritization matrix generated in Phase 1 (See Table 3).

PAGE 43

28 Phase 4 Project prioritization The combined database containing each prioritization element (main breaks, service main breaks, etc.) for each planned cap ital project is then evaluated from the perspective of projected financ ial benefit of proceeding with each planned capital project. The prioritization matrix developed in Phase 1 is applied to determine a cost for each occurrence of a priori tization element, or benefit unit The benefit unit for each master plan program is determined based on th e developed prioritization matrix and the prioritization element having the greatest tang ible value. A table of benefit units is created as shown in Table 5 Table 5. Overlay of benefit units of each priori tization element with master plan combinations to allow for calculation of benefit to cost ratios. Cost of Prioritization Elem ent Occurrences (Benefit Units) Master Plan Program Water Main Breaks Service Main Breaks Quantity of Proposed New Hydrants Water Age Quantity of Flushing Visits Water Pressure Complaints Water Quality Complaints Hydraulic Looping [$ per break] [$ per break] [$ per hydrant] [$ per age increment] [$ per visit] [$ per complaint] [$ per complaint] Undersized Mains [$ per break] [$ per break] [$ per hydrant] [$ per age increment] [$ per visit] [$ per complaint] [$ per complaint] Unlined Cast Iron [$ per break] [$ per break] [$ per hydrant] [$ per age increment] [$ per visit] [$ per complaint] [$ per complaint] Asbestos Cement [$ per break] [$ per break] [$ per hydrant] [$ per age increment] [$ per visit] [$ per complaint] [$ per complaint] The net present value (NPV) of project ed prioritization element occurrences during the next 20 year planni ng period is computed based on an assumed discount rate

PAGE 44

29 and a projected rate in water main constr uction cost increases. Predicting future occurrences is accomplished by determining the average annual increase of prioritization element occurrences for all planned capital projects. The average increase is then multiplied by the number of years into the future, and added to the historical average number of occurrences for each planned capital project. If sufficient historical data are available, an individual projection can be applied to each planned capital project to compensate for errors associated with the li near extrapolation that may not accurately reflect the expected times a pa rticular asset will fail. More rigorous models predicting pipe failure can also be incor porated into this project at this step in the prioritization process. The net present value of the benef it associated with preventing the occurrences of a prioritization element is then determined. Equation (1) can be used as a method to determine the tota l benefit of each prioritization element. j ring year urences du ment y occ zation ele f prioriti quantity o y N occurence dollars ] gram x, [ r plan pro y in maste n element oritizatio it for pri benefit un x y BU ven policy dri ogram x er plan pr y in mast on element ioritizati ctor of pr benefit fa x y k ] dollars [ project i ent y for ation elem prioritiz benefit of i y B where yrs n yrs j y N x y BU NPV x y k i y B , (1) 0 )} ( {

PAGE 45

30 Once the individual benefits of each prioriti zation element are calculated, a net benefit factor can be calculated for each project from Equation (2). ect i it of proj Net benef i B where all y i y B i B (2) Based on the estimated cost per foot, the construction cost of each planned capital project, Ci, is then estimated. Once the net benefit and construction cost for each planned improvement project are determined, a ratio can be established to prioritize projects as defined in Equation (3). (3) i C i B Ratio tion Prioritiza The database is then sorted by the priori tization ratio to obtain a prioritized list of planned capital projects.

PAGE 46

31 RESULTS Results for each of the four project ph ases are included below. Generally, the established methodology shows to be effectiv e for prioritizing pla nned capital projects within the TWD service area. Phase 1 Policy decisions A survey conducted among Tampa Water Department policy makers including the director, chief engineer chief design engineer, chief planning engineer, and a planning engineer determines benefit factors to be applied in a cost to benefit prioritization methodology. Survey results are shown in Table 6. Table 6. Results of Tampa Water Department su rvey to determine benefit factor values. Master Plan Program Water Main Break Service Main Break Quantity of Proposed New Hydrants Modeled Water Age Quantity of Flushing Visits Water Pressure Complaint Water Quality Complaint Hydraulic Looping 10% 70% 10% 10% Undersized Mains 30% 10% 50% 10% Unlined Cast Iron 50% 10% 10% 20% 10% Asbestos Cement 80% 10% 10%

PAGE 47

32 Phase 2 Data collection and construction costs The specified methodology for collecting da ta and establishing construction costs adequately describes the steps necessary to allow for later GIS analysis. Results of the data collection and the determination of estim ated capital costs are described below. Data collection Planned rehabilitation and replacement pr ojects are successfu lly converted to a geospatial format utilizing the USA NAD83, Flor ida State Plane, West Zone coordinate system. This conversion involves 2,886 proj ects consisting of 495 miles of planned water main improvements. Attribute data including the master plan program, proposed nominal pipe diameter, street in which the project is located, beginning and ending streets defining the project boundary, re ference atlas page, and the ap proximate water age at the project location are incorporated for each pla nned project. The tota l length of planned rehabilitation, replacement and improvement projects for each master plan category is shown in Figure 8.

PAGE 48

33 Total Length of Rehabilitation & Replacement Projects by Type (miles) A CMR, 16 UMR, 346 UCIMR, 114 HLS, 19 Figure 8. Pie chart comparin g quantity of planned TWD rehabilitation, replacement, and improvement projects (HLS = hydraulic looping system, ACMR = asbestos cement main replacement, UMR = undersized main replacement, UCIMR = unlined cast iron main replacement) The relative locations of pla nned capital improvement projec ts are shown in Figure 9.

PAGE 49

34 Figure 9. Relative location of various types of planned rehabilitation, replacement and improvement projects within the TWD service area. Prioritization elements specified in Table 6 are also successfully compiled. This information is presented in Figures 10 through 16.

PAGE 50

35 Figure 10. Water main break locations within the TWD service area.

PAGE 51

36 Figure 11. Service main break locations within the TWD service area.

PAGE 52

37 Figure 12. Water main flushing locati ons within the TWD service area.

PAGE 53

38 Figure 13. Water quality complaint loca tions within the TWD service area.

PAGE 54

39 Figure 14. Water pressure complaint locations within the TWD service area.

PAGE 55

40 Figure 15. Proposed fire hydrant locations within the TWD servi ce area. A total of 2,607 fire hydrant locations were id entified that require a water main improvement. A total of 2,584 fire hydrant locations were identi fied that do not require a water main improvement.

PAGE 56

41 Figure 16. Results of water age modeling within the TWD service area.

PAGE 57

42 Construction costs Estimated construction cost factor s are obtained from the Tampa Water Department. Meter spacing is assumed to be 0.03 meters per foot and the cost of a meter set is determined to be $1,411. Cost factors for various nominal pipe diameters utilized for this project are provided in Table 7. Table 7. Cost factors utilized for various nominal pipe diameters ranging from 4-36 inches Nominal Diameter (in) Cost* ($/ft) 4 83 6 96 8 102 12 144 16 200 20 224 24 295 30 479 36 517 *Per foot costs having a nominal diameter greater than 12 is based on a linear extrapolation.

PAGE 58

43 Attribute data for each planned rehabili tation and replacement project is exported into a spreadsheet wher e the above cost factors are applie d. The total construction cost of all planned rehabilitation or replacement pr ojects is estimated to be $388 million. The current construction cost of all planned reha bilitation or replacem ent projects by master plan program is illustrated in Figure 17. Planned CIP Costs by Project Type ACMR, $12,500,000 UMR, $262,700,000 UCIMR, $98,500,000 HLS, $14,500,000 Figure 17. Estimated construction costs for planned TWD rehabilitation, replacement and improvement projects (HLS = hydraulic looping system, ACMR = asbestos cement main replacement, UMR = undersized main replacement, UCIMR = unlined cast iron main replacement)

PAGE 59

44 Phase 3 Data extraction The structured query presented in th e methodology section is shown to be effective at linking occurrences of prioritiza tion elements to planned capital projects. Once these queries are performed, a single database is developed to aid in prioritization. Structured queries Buffer region dimensions are determined based on typical lot sizes and other previously discussed factors. The buffer dimens ions determined to be appropriate for this project are shown in Table 8. Table 8. Dimensions of buff er regions for each type of prioritization element. Prioritization Element Buffer Region Radius Dimension (feet) Water Main Breaks 125 Service Main Breaks 125 Proposed New Hydrants 125 Water Age 125 Flushing Visits 450 Pressure Complaints 125 Water Quality Complaints 125 Created planned project buffers with the lo cation of prioritization element occurrences are shown in Figure 18.

PAGE 60

45 Figure 18. Visual representation of planned project buffer regions and prioritization elements Database development Once the SQL queries are performed with in MapInfo, a database containing all planned rehabilitation and repl acement projects and all prior itization elements within the respective buffer region is generated by u tilizing the unique id entifying key. This database is used to determine average occurr ences of various prioritization elements for fiscal year 2002 through fiscal year 2005. Results are summarized in Figure 19.

PAGE 61

46 Prioritization Element Occurences0 0.05 0.1 0.15 0.2 0.25 FY-02FY-03FY-04FY-05Events per 1,000 ft of Water Main Water Main Breaks / 1000 ft Service Line Breaks / 1000 ft Pressure Complaints /1000 ft Quality Complaints / 1000 ft Figure 19. Comparison of average quantities of prioritization element occurrences among planned rehabilitation, replacement and improvement projects The data are further digested to yield a hist orical annual average in crease or decrease in the occurrences of various prioritization elem ents. A summary of these infrastructure condition indicators is shown in Table 9.

PAGE 62

47Table 9. Relative annual changes in water main breaks, service line breaks, pressure complaints an d water quality complaints Water Main Breaks Service Line Breaks Pressure Complaints Water Quality Complaints FY02-FY03 5% 1% 16% -128% FY03-FY04 24% -29% -23% 51% FY04-FY05 16% 61% 7% -199% AVG 15% 11% 0% -92% These infrastructure condition indicators are used as factors to forecast prioritization element occurrences for the next 20 year period. Phase 4 Project prioritiza tion and master planning Prioritization element occu rrences are converted to monetary costs based on the benefits specified in Table 6. It was assumed the total cost of a main break is $1,500 and the cost of a water main flushing activity is $100. The resultant co sts to the Tampa Water Department based on developed benef it units are listed in Table 10. Table 10. Benefit unit based costs ut ilized in the prioritization process Program Water Main Breaks Service Line Breaks Quantity of Proposed New Hydrants Modeled Water Age Flushing Time Water Pressure Complaints Water Quality Complaints Hydraulic Looping System Improvements $1 each increment $70 / flush $10 $10 Fire Protection / Undersized Main Replacement $450 $150 $750 $150 Water Main Rehabilitation or Replacement $750 $150 $150 $300 $150 Cement Main Replacement $1,200 $150 $150

PAGE 63

48 These costs are then applied to the deve loped database containing the historic and projected occurrences of prioritization elemen ts. Utilizing a 3 percent cost escalation for expected costs and a 7 percent discount rate, a net present va lue of all costs associated with maintaining each water main planned to be rehabilitated or replaced is calculated for a 20 year planning horizon. The constructi on cost of each planned replacement or rehabilitation project is then combined with the net present value of expected benefits to obtain prioritization ratios. The resulting ratios vary from 0.35 to 0.00. However, the distribution of the results show a relativel y small quantity of planned projects having a benefit to cost ratio greater than 0.1. Figure 20 shows the approximate benefit to cost ratio distribution. Benefit to Cost Ratio Distribution 63.8% 29.3% 4.7% 1.2% 1.0% 0.000 0.025 0.025 0.050 0.050 0.075 0.075 0.100 0.100 1.000 Figure 20. Results yield benefit to cost ratios ranging from 0.35 to 0.00. A relatively small portion of projects have benefit to cost ratios greater than 0.10.

PAGE 64

49 DISCUSSION The use of GIS permits the developed prioritization system to analyze large numbers of planned capital projec ts to efficiently generate benefit to cost ratios. This system has several key benefits including short and long term pl anning, the ability to monitor the general condition of water main in frastructure and to a ppropriately allocate available funding resources among master plan programs. Prioritization system Allowing an efficient method for associat ing prioritization elements to planned projects, buffer zones can be utilized with reasonable accuracy when resources do not allow for the full scale implementation of a computerized maintenance management system. Although the use of GIS software and the interpretati on of data requires specialized personnel, large numbers of personnel do not need to be trained on new software packages such as computerized maintenance management systems.

PAGE 65

50 Integrated database Combining rehabilitation, replacement and improvement master plans into a single geospatial database provides a means to improve access to data and increase the ease of use. The availability of this information cleari nghouse has signifi cant long-term benefits for the water utility. Some questions that can be answered using the integrated database concepts develope d in this project are: 1. Which distribution system projects ma ximize available resources during a given budget year? 2. What is the condition of the infrastruc ture near a planned development and what planned infrastructure improvements should the developer be required to fund? 3. What is the 5 and 20 year plan for infrastructure rehabilitation, replacement and improvements? The developed database can also be utilized as part of a futu re integrated infrastructure management system.

PAGE 66

51 Database applications Depending on the size of the area to be analyzed, answering questions regarding planned infrastructure in a specific service area section historically requires extensive soft resources and can not be reas onably performed in the time frame of a few hours due to inefficient methods of storing data. The electronic data fo rmat created by this project reduces the soft resources required to answ er common questions such as what are the planned projects, when are they scheduled to be completed and how much will they cost. The resources necessary to answer these que stions has been reduced by an estimated 85 percent. For example, an analysis of pla nned capital projects w ithin a recent proposed development area consumed the time of a sta ff engineer and one e ngineering technician for approximately three days. Once the methods described in this proj ect were applied to the same area, the same task could be comp leted in approximately 4 hours primarily by an engineering technician. Condition indicators Between FY2002 and FY2005 the TWD has implemented projects such as ozonation that have decreased water quality complaints near planned distribution system projects by an average of 92 percent annuall y. Replacing unlined cast iron water mains and installing hydraulic looping system project s are expected to continue decreasing the quantity of water quality compla ints in these areas.

PAGE 67

52 However, the presented analysis of wa ter main breaks and service main breaks near planned rehabilitation, replacement and improvement projects indicate the condition of water infrastructure near th ese planned project locations is gradually deteriorating. Between FY2002 and FY2005, water main breaks have increased near planned project locations by 15 percent annually while service main breaks near the same locations have increased by 11 percent annually. These trends are consistent with the national issue of deteriorating utility infrastructure within the United States. It is expected utilizing the prioritization technique presented in this pr oject will decrease the realized effects of deteriorating infrastructure by replacing the pi pe segments in the worst condition first. Resource management Utilizing the developed prioritization techni que, the distribution of benefit to cost ratios is skewed strongly left, indicating a relatively small number of projects are more favorable for implementation than others. This suggests relatively small resource expenditures can generate a relatively la rge amount of value when resources are efficiently utilized. Currently, the TWD c hooses all projects w ithin a neighborhood and constructs them during the relatively same ti me period. However, at the current funding levels, TWD is typically not able to replace or rehabilitate all pla nned projects within a neighborhood. Therefore, regardless if the pr ojects are more spread out throughout the service area, it is more appropriate to select the most cost effective projects rather than pursuing a neighborhood-wide approach.

PAGE 68

53 Utilizing the data obtained in this project funds can be distributed appropriately between each master plan program. Analyzing the top 10% of planned capital projects, it is shown the current rehabilitation and repl acement budget distributi on can be improved. The undersized main replacement funding should be approximately four to five times that of unlined cast iron replacement funding. Figure 21 shows the existing and proposed funding levels for each master plan program. Projected Cost Distribution for Top 10% of Projects 76.3% 16.2% 6.6% 0.9% Current Budget Distribution for Rehabilitation & Replacement42.6% 42.6% 10.6% 4.3% UMR UCIMR ACMR HLS Figure 21. Comparison of existing rehabilitat ion and replacement funding levels and the ideal funding levels as determined by the top 10% of prioritized projects determined by this project. The obtained data can also be utilized to determine if utility rates are adequate to sustain existing service levels. Because th e quantity of water main breaks and service line breaks presented in this project are increasing, it can be implied existing rehabilitation and replacement funding levels should be increased to sustain service levels. The necessary rate increase amount is beyond the scope of this project.

PAGE 69

54 CONCLUSION This project successfully c onsolidates master plan programs, main break reports, service main break reports, customer comp laint logs, flushing reports, fire hydrant spacing criteria and estimated water age into a single database eliminating analysis difficulties due to inconsistent data formats. Once a consolidated database is obtained, benefit to cost ratios for 2,886 planned projects are successfully calculated based on GIS analysis. Analyzing the top 10 percent of plan ned capital projects allows the water utility to appropriately allocate av ailable funding between master plan programs. Specific conclusions for each project objectiv e are listed below. 1. Objective: Convert existing infras tructure replacement master plans into an electronic format incor porating attribute data to permit geospatial analysis. Conclusion: Master plan programs are successfully converted into an electronic format appropriate for GIS analysis. 2. Objective: Consolidate existing databases including main break reports, service main break reports, customer complaint logs, and flushing reports into a common form at that can be utilized by GIS software.

PAGE 70

55 Conclusion: Databases containing listed infrastructure condition indicators are consolidated into a common format suitable for GIS analysis. 3. Objective: Perform geospatial analys es to determine a benefit to cost ratio for each planned project based on a prioritization matrix agreed upon by Tampa Water Department policy makers. Conclusion: A prioritization ma trix developed by Tampa Water Department policy makers is develo ped. Benefit to cost ratios for 2,886 planned capital projects are included as part of this project. 4. Objective: Prepare a combined, pr ioritized infrastructure replacement master plan and make appropr iate budget recommendations. Conclusion: A prioritized capital project master plan and budget recommendations are made as part of this project. The prioritization procedure specified in this project is shown to be a viable alternative for utilities already owning licenses to GI S software and do not currently have the benefits of full-scale computerized maintenance management systems available for their use.

PAGE 71

56 ENGINEERING IMPLICATIONS Beneficial use of the prioritization st rategy developed by this project depends on procedural changes to maintain necessary da tabases as well as the use of engineering judgment to interpret project results. Proposed procedural changes are included and the need to still use common sense judgment for individual projects is addressed below. Procedural changes The prioritization system developed in this project can provide a robust tool that utilities can use to modify existing procedures to maintain geospatial data in a useable, up to date format. Maintaining geospatial data requires i nput from personnel involved in engineering, line maintenance, information technology, customer service and records. Project planning is a dynamic process that needs to respond to ongoing occurrences of water main breaks, service line breaks, flushi ng activities and custom er complaints. In addition, the inventory of planned projects is continually changing due to redevelopment and urban infill. The implementation of planned replacement or rehabilitation projects is often due to the need for additional ca pacity and reliability demanded by these redevelopment and urban infill projects, not necessarily because they were listed as a potential capital project. These different dr iving forces demonstrate the need to link

PAGE 72

57 rehabilitation and replacement planning w ith infrastructure upgrades necessary for redevelopment and urban infill projects. A conceptual procedure to continually update the inventory of planned projects when capital improvement projects and customer driven projects occur simultaneously is shown in Figure 22.

PAGE 73

58 Figure 22. Procedures are needed to conti nually maintain geospatial databases when customer driven projects and capital project s occur simultaneously. The process flow diagram above shows input from several functional groups are needed to accomplish this task.

PAGE 74

59 Improved resources to support decisions The procedures developed by this pr oject provide a valuable resource for prioritizing planned water infrastructure projects. One should not be 100 percent replaced by the procedures de veloped in this project. The ready availability of this information coupled with the opportunity for scenario analysis can aid engineers and planners in making sound j udgments pertaining to infras tructure projects.

PAGE 75

60 ADDITIONAL RESEARCH Although the prioritization tec hnique presented in this project is shown to be successful, key questions remain unanswered for the Tampa Water Department and are included below as recommendations for additional research. 1. How much can implementation costs be reduced by right-sizing projects? During this project, it became known addi tional work is needed on the undersized main replacement (UMR) master plan. Nearly all planned UMR projects are proposed as 8-inch diameter improvements. A large porti on of these projects can be reduced in size to 6-inch diameter improvements and still comply with TWD fire flow standards. Downsizing UMR proposed pipe sizes could pot entially reduce capital expenditures by 6 percent based on the difference between 8-inch and 6-inch pipe installation costs. Because the cost of the project is used to pr ioritize the projects, this should be done on a system wide basis rather than the curren t method of downsizing the project as it is selected for design. 2. Can buffer regions be improved increasing project accuracy? One significant improvement that can be made to the method presented in this project is to eliminate inaccura cies associated with assigni ng buffer regions to associate

PAGE 76

61 planned projects with prioritization elemen t occurrences such as main breaks and customer complaints. Figure 23 illustrates one weakness that results from buffer regions. Figure 23. Example of inaccurate assignments of prioritization elements due to the use of buffer regions. One way to eliminate this error is to develop a full scale computerized maintenance management system for all distribution piping operated by the TWD. This system should incorporate a geospatial component so reco rds of prioritization occurrences can be associated with the appropriate segment of piping without error. The benefit to cost

PAGE 77

62 analysis utilized in this projec t is still effective, however the method to acquire the input data in Phase 3 would be greatly increased in accuracy. 3. What additional efficiency can be realized by developing an integrated infrastructure management system? Opportunities exist to optimize capital expenditures by expanding the prioritization process to include other infrastructure types su ch as storm water, sanitary sewer, sidewalks, and pavement. The tech niques developed for this project could be carried over into an integrated infrastructu re management system provided the necessary interdepartmental coor dination occurs. 4. What savings would result from incorp orating pipe age and improved life cycle models into the prioritization process? Pipe age can be a very good general indicator of infras tructure condition. Incorporating this into this project is curre ntly not practical due to the difficulty in accessing historical records, however pipe installation dates should be included as attribute data associated with all pipe currently being inst alled to facilitate similar projects in the generations to come. 5. Can currently available techniques suffici ently prioritize transmission and grid system projects?

PAGE 78

63 The TWD delivery and grid system master plan should be prioritized. One way to accomplish this task is through genetic algor ithm optimization utilizing existing hydraulic models. Genetic algorithm optimization is a se ries of computer calculations designed to iteratively generate optimal implementation scen arios. Questions that can potentially be answered by this type of computer modeling include which projects should be selected, what size pipes should be installed and which projects should be constructed together to obtain synergistic effects. Genetic algorithm optimization may also incorporate a master plan for expanding and optimizing wate r storage and pumping sequences. 6. What is the Tampa Water Depar tment infrastructure budget Gap? The difference between the current rate of infrastructure reinvestment needs to be compared against the rate necessa ry to maintain the existing le vel of service. This would assist in making budget recommendations as well as justify any necessary rate increases. 7. Is pipe rehabilitation a viable option for the Tampa Water Department? The TWD relies on pipe replacement in stead of pipe rehabilitation for infrastructure projects. A study comparing the economics of the various options should be performed to further optimize the use of capital resources.

PAGE 79

64 REFERENCES American Water Works Association, presented by Hooker, Michael. Statement on the Fiscal Year 2003 budget of the EPA, testimony to congress before the Subcommittee on Vulnerability Assesment, HUD, and Indepe ndent Agencies of the Committee on Appropriations. (April 16, 2002). Washington D.C. American Water Works Associatio n, presented by Neukrug, Howard. Statement on Drinking Water Needs and Infrastructure before the Environment and Hazardous Materials Subcommittee, Committee on Energy and Commerce, U.S. House of Representatives. (March 28, 2001). Washington D.C. American Water Works Associat ion, presented by Neukrug, Howard. Statement on Drinking Water Needs and Infrastructure before the Environment and Hazardous Materials Subcommittee, Committee on Energy and Commerce, U.S. House of Representatives. (April 11, 2002). Washington D.C. Anderson, D. and Smith M. Common Sense and Asset Management Accessed May 2006. http://www.cdm.com/NR/rdonlyres/A4FDD508-05B9-463C-B666B4FA6B82471D/0/CommonSen eseandAssetManagement.pdf Congressional Budget Office. Future Investment in Drin king Water and Wastewater Infrastructure. (2002). Washington, D.C. Ferreira, A. and Duarte, A. A GIS-Based Integrated Infrastructure Management System. (2005). FIG. Portugal. Grablutz, F. and Hanneken, S. Economic Modeling for Prioritizing Pipe Replacement Programs. (2001). Roy F. Weston, Inc. and St. Loui s County Water Company. St. West Chester, PA and St. Louis, MO. H2O Coalition, presented by Beecher, Janice A. (March 28, 2001). Testimony on Water Infrastructure Needs before The House Committee on Energey and Commerce. Washington D.C.

PAGE 80

65 Lemer, A.C., Ph.D. Progress Toward Integrated In frastructure-Assets-Management System: GIS and Beyond. (1998). APWA International Pu blic Works Congress. Las Vegas, NV. Lim, E. and Pratti, R. Pipe Evaluation System (PIPES). (1997). Seattle Public Utilities. Seattle, WA. Martin, Terry. Modeling System Leverages GI S to Assess Critical Assets. (April 2005). WEFTEC WaterWorld. Tulsa, OK Matichich, M., Allen, J., and Allen R. Asset Management Planning. (2006). Journal AWWA, Vol. 98, Number 1, pp 80-87. Denver, CO. Misha, H. Government Services, Inc. Costs of Infrastructure Failure. (2002). American Water Works Association Research Foundation. Nagel, R., and Elenbass, M. Prioritizing Capital Improvemen t Projects to Mitigate Risk. 2006). Journal AWWA, Vol. 98, Nu mber 1, pp 72-79. Denver, CO. U.S. Environmental Protection Agency. The Clean Water and Drinking Water Infrastructure Gap Analysis. (2002). Office of Water. Washington, D.C. U.S. Environmental Protection Agency. Drinking Water Infrastructure Needs Survey First Report to Congress. (1997). Office of Ground Wa ter and Drinking Water. Washington D.C. U.S. Environmental Protection Agency. Drinking Water Infrastructure Needs Survey Second Report to Congress. (2001). Office of Ground Wate r and Drinking Water. Washington D.C. U.S. General Accounting Office. Water InfrastructureComprehensive Asset Management Has Potential to Help Utilitie s Better Identify Needs and Plan Future Investment, Report to the Ranking Minor ity Member, Committee on Environment and Public Works, U.S. Senate. (2004). Washington, D.C. U.S. Government Affairs Office and th e American Water Works Association. Dawn of the Replacement Era, Reinvesting in Drinking Water Infrastructure. (2001). Washington, D.C. Vanier, D., Danylo, N., and Ville de Montreal Finance Department. Municipal Infrastructure Investment Planning: Asset Management (1998). APWA International Public Works Congress. Las Vegas, NV.

PAGE 81

66 Weston, contractor. Decision Support System for Distri bution System Piping Renewal. (2002). American Water Works Association Research Foundation. Zhang, J, Project Manager Co lorado State University. Assessmment and Renewal of Water Distribution Systems American Water Works Association Research Foundation.

PAGE 82

67 BIBLIOGRAPHY Brown, Jeanette. Water Infrastructure Earns Din ASCE Report. (April 2005). WEFTEC WaterWorld. Tulsa, OK. Deb, A., Kerz, R., Hasit, Y, and Grablutz, F. Quantifying Future Rehabilitation and Replacement Needs of Water Mains. (1998). American Water Works Association Research Foundation. Denver, CO. Engineering Department. Deteriorating Buried Infrastr ucture Management Challenges and Strategies. (2002). American Water Works Serv ice Co., Inc. Voorhees, N.J. Gilmore, W., Fairchild, P., and Kellogg, S. Managing Asset Management. (Accessed March 2006) http://www.cdm.com/NR/rdonlyres/C3938E5D-A589-4182-8EVC88E38EED4828/0/ManagingAssetManagement.pdf. Cambridge, Massachusetts. Hohn, C. and Payne, M. GIS Improves South Dakota Water. (2006). American Water Works Association Opflow, Vol. 32, No. 1, pp 1, 4-7. Denver, CO. Laughlin, James, Ed. EPA Report to Congress Examines US Drinking Water System Needs. (September 2005). WEFTEC WaterWorld. Tulsa, OK. Means, E., Bernosky, J., and Patrick, R. Technology Trends and Their Implications for Water Utilities. (2006). Journal AWWA, Vol. 98, Number 1, pp 60-71. Denver, CO. Miner, Gary, Ed. Water Industry Realizing Benefits From GIS Applications. (2005). Journal AWWA. Denver, CO. Petrescu, Florian. Teaching GIS to Urban Engineers. (2004). 10th EC GI & GIS Workshop, ESDI State of the Art. Warsaw, Poland. Smith, M., Lovett, M., and Caldwell, J. Designing an Asset Management System. (Accessed March 2006). www.cdm.com U.S. Department of Transportation. (1999). Asset Management Primer. Federal Highway Administration, Office of Asset Management.

PAGE 83

68 U.S. Environmental Protection Agency. (2002). Asset Management for Sewer Collection Systems FACT SHEET. Office of Wastewater Mana gement. Washington D.C. U.S. General Accounting Office. Water Infrastructure, Information on Financing, Capital Planning, and Privatization, Re port to Congressional Requesters. (2002). Washington, D.C. Waldron, K. and Ratchinsky, W. Sewer System Evaluation Study (SSES) Utilizing GIS Tools. (1997). APWA Seminar Innovations in Urban Infrastructure. International Public Works Congress. Minneapolis, MN. Water Infrastructure Network. Clean & Safe Water for the 21st Century, A Renewed National Commitment to Water and Wastewater Infrastructure. (2001). Washington, D.C. Water Infrastructure Network. Water Infras tructure, Recommendations for Clean & Safe Water in the 21st Century. (2001). Washington, D.C.