RP015-23 Ad#1

March 28, 2023

ADDENDUM #1

RP015-23, Provision of Advanced Metering Infrastructure (AMI) System

Requirements, Implementation Plan and RFP Development

**THE PROPOSAL DEADLINE HAS BEEN EXTENDED TO 2:50 P.M. ON APRIL 25,2023.**

*QUESTION DEADLINE IS EXTENDED TO 10:00 A.M. ON APRIL 4, 2023.*

This Addendum is made part of the Proposal Documents of the above noted solicitation.

QUESTIONS AND ANSWERS:

1. Please provide a copy of the prior AMI Study conducted for Gwinnett County. See attached.

2. Has the County selected a firm already? No

3. Will the County need the consultant to assist with RFP development only or will they be assisting with the evaluation process as well? The awarded firm will assist with the RFP development only. They may also provide advisement to assist in negotiations with the highest scoring firm, but Gwinnett County will be responsible for negotiations.

4. “Assist in Proposal Evaluation Process” is Item 5 on the Cost Proposal. Can you confirm that Proposal Evaluation Support is within the scope of this RFP? Yes, the vendor will provide subject matter expert support during the proposal evaluation. The vendor will not be part of the scoring committee, but instead provide review criteria focused on what to look for in a successful proposal.

5. “Negotiation support” is included as Item f on page 4 of the RFP; however, it is not listed on the Cost Proposal. Can you confirm that Negotiation Support is within the scope of this RFP?

Negotiation Support will not be part of the scope under this contract. Any reference to this being a requirement shall be removed.

6. On page 4 of the RFP, a “Recommendation report” (Item g.) is included for Task 2:

Procurement. It appears this report will be produced after contract negotiations. What is the anticipated content and who is the audience for the Recommendation Report? Any reference to this being a requirement shall be removed.

7. Would the County be willing to accept digital submissions? All submittals must be delivered in a sealed envelope per the RFP. Digital responses will not be opened or considered.

8. The RFP has indicated a 6-month timeline, is this just for Phase 1? Our experience with this type of work would indicate this timeline is aggressive, would the County be willing to accept timelines that are more reflective of the workload requested. The project timeline is for the entire project and shall remain as stated in the RFP. Proposals should reflect this timeline.

RP015-23 Ad#1

REVISIONS:

1. Please revise the following language on Page 1 of the Invitation to read; ”One unbound single-sided original, five (5) exact copies, and one (1) identical electronic pdf copy on flash drive should be submitted. Do NOT include cost information on the flash drive.”

NOTES:

1. Please be sure the contact information is accurate and up to date on the Reference Sheet. All proposers are strongly encouraged to let those contacts know that the County will be reaching out to them. This will help the County avoid delays in the evaluation process.

2. Insurance Requirements are listed in the Invitation. Insurance does not have to be submitted with your proposal, but will be required of the awarded firm. Proposers should contact their insurance provider to ensure the required coverages are met. The County will not pay additional costs not included in the Cost Proposal.

3. Proposers are not to submit pricing outside of the line items requested in the Cost Proposal.

Any questions, comments or concerns regarding the structure of the Cost Proposal, specifications or anything else included in the RFP should be directed to Kaley Ivins. If the County receives information and determines a change needs to be made, that information will be issued in an addendum.

Acknowledge receipt of this addendum on the Consultant Information Form (Page 9) of the proposal document.

Thank you, Kaley Ivins, CPPB Purchasing Manager

Attachments:

1. Pre-Proposal Conference Attendance Sheet

2. AMI Study

T E C H N I C A L M E M O R A N D U M

ADVANCED METER INFRASTRUCTURE

FEASIBILITY STUDY

FINAL REPORT

Gwinnett County Department of Water Resources

08 May 2019

Contents

EXECUTIVE SUMMARY

Key Findings

Business Case Results

Additional Benefits

Recommendations

COST BENEFIT ANALYSIS

Objectives of the Model

Assumptions

Scenario Definitions

Cost/Benefit Categories Positive Financial Benefits Additional Revenue Cost Savings Avoided Costs Negative Financial Impacts Additional Operating Costs

Economic Evaluation Model Results

Other Benefits and Opportunities

CONCLUSION

Path Forward Define Key Success Criteria and Measurement Strategy Proceed to Procurement

APPENDIX A: GLOSSARY OF TERMS

Executive Summary The Gwinnett County Department of Water Resources (DWR) is investigating the feasibility of a system-wide deployment of Advanced Metering Infrastructure (AMI) to upgrade its meter reading and other operational processes to provide continued best-in-class service to its residential and commercial customers. DWR has already implemented numerous business and customer-service policies that put it well ahead of industry standards; however, it is constrained in its ability to offer certain customer-valued services that can only be enabled by AMI. DWR contracted with Jacobs Engineering Group (“Jacobs”) to perform a Feasibility Study to gain an understanding of current technology options and to explore the benefits of a full AMI implementation for DWR and its customers.

The intention of this report is to present the results of the feasibility study that Jacobs has conducted on behalf of DWR. A separate report has been submitted regarding DWR’s meter inventory analysis and initial budgetary cost estimates for AMI. This report identifies and, where possible, quantifies the benefits of an AMI system to DWR, provides updates to budgetary cost information to reflect market changes occurring since the previous report was delivered, and calculates a 20-year cost-benefit analysis to evaluate the program’s financial feasibility. It also details information DWR can use to improve its operations, reduce costs, enhance customer benefits, and better serve the residents of the County.

Key Findings Key findings of the feasibility study include the following.

• The average throughput from existing meters is 495,000 gallons, indicating the existing meter populations is on average half-way through its useful life, and based on meter test data, has slowed to a point where more than 300 million gallons of apparent losses may be incurred each year due to meter under-registration.

• Approximately 11% of the existing residential meter population is at or above one million gallons of total throughput and should be replaced as soon as possible.

• Converting from visual meter reading to AMI could result in a significantly lower cost (average savings of $230 per customer over twenty years) compared to the current contracted prices for meter replacement and meter reading costs, based on assumptions and estimated developed in this BCE.

• The level of effort that would be needed to address normal and anticipated AMI system maintenance (radio issues, broken meters, etc.) is equivalent to the level of effort currently expended by DWR on the existing system, based on assumptions and estimates developed in this BCE.

• AMI would eliminate nearly 90% of the current Bermex services. It is expected that the unit cost for the remaining services will increase substantially and could be replaced more cost effectively by new internal DWR hires.

Business Case Results In addition to the above key findings, Jacobs has completed a full cost-benefit analysis. The results provide a very strong financial justification for AMI. Some of the key outcomes are highlighted below:

• The cost to continue meter replacement and reading activities under DWR’s existing contracts for the next 20 years is $24 million.

• The Net Present Value (NPV) cost of an AMI implementation including 20 years of system maintenance and network costs is $58 million for a fixed network and $65 million for cellular.

• Additional value drivers for AMI include $7 million in improved meter registration and $31 million in savings from the Bermex contract.

The overall results (expected value) of AMI compared against the Base Case are shown in Figure 1 and indicate that a full AMI deployment would provide DWR with an average annual benefit of at least $3.7 million, a payback of between 12 and 14 years, and a return on investment of over 7%. Both network alternatives have a positive NPV, which indicates the benefits of the project over 20 years would exceed the cost of financing, even if DWR were to sell a bond to fund the program.

It should be noted that the below calculations reflect a scenario where DWR either eliminates the Bermex contract and hires new in-house staff or negotiates the unit prices of a reduced Bermex contract to not exceed +75% of their current value. This is the conservative scenario. Should DWR negotiate a more favorable contract or require fewer new staff than anticipated, these metrics will be even stronger. Jacobs has included toggles in the Business Case Model to assess how these metrics change under a variety of Bermex scenarios.

AMI Business Case Outputs: Fixed Network Cellular Base Case

Initial Capital Cost $ 52,660,000 $ 51,360,000 $ -

Average Annualized Benefit $ 4,465,000 $ 3,790,000 $ -

Simple Payback (years) 12 14 0

Net Present Value $ 8,570,000 $ 900,000 $ (24,070,000)

Return on Investment (ROI) 8.0% 7.0% 0.0%

Figure 1: Overall Results – Comparison between two AMI scenarios and the Base Case

Additional Benefits In addition to the financial justification for the project, there are several other benefits which have not been quantified, but that DWR will realize by implementing a system-wide, consolidated AMI system.

For some utilities, these non-quantified benefits are as much of a justification for the project as the financial drivers. These benefits include:

• Improved Customer Service – Full AMI implementation will help DWR significantly improve customer service. Bill disputes will drop in frequency and will be easier to resolve when they do occur. Leaks can be identified and addressed expeditiously. Not only will AMI enable DWR staff to provide a richer customer service experience, but it will also improve the job satisfaction of customer service representatives by providing them with the tools and data to do their job more effectively in helping to answer customer questions. And within an environment of rising rates, AMI is a tool that can be used to empower customers to manage their water use and related costs.

• Environmental Benefits - AMI will also soften DWR’s environmental impact by enabling it to reduce greenhouse gases, facilitate water conservation and water efficiency (including enhanced ability to manage water loss using DMAs), detect backflows, and reduce energy use for pumping.

• Operational Benefits – While Jacobs has captured some operational savings in the benefit cost model, others are less definitive and may require specific process improvements to implement.

Typically, utilities often find significant value through standardization of systems and software, improved drought management, improved route management, improved health and safety, hydraulic model calibration, improved cost of service analysis for ratemaking, and right sizing of large meters.

Recommendations The cost-benefit analysis, as detailed in this report, provides strong financial justification for deploying

AMI system-wide. Due to the positive outcomes of this comprehensive analysis, it is recommended that

DWR pursue a systematic deployment of AMI. To make sure that DWR can capture and be able to demonstrate desired outcomes from a system-wide deployment, Jacobs recommends DWR take the following next steps:

• Identify and establish the means to measure progress towards key success criteria;

• Develop a communication plan for both internal and external stakeholders based on the identified key success factors; and

• Develop an organizational action plan that includes customer outreach activities, skills training, internal communication, changes in position responsibilities, business process changes, and system integration

Once DWR is ready to proceed (either in parallel with the above, or after the above actions), it is recommended that DWR develop a performance-based procurement document and solicit competitive technical and cost proposals for a holistic AMI solution to be deployed within 3-4 years.

Cost Benefit Analysis Jacobs has developed an Excel-based business case model to support the cost-benefit analysis, calculate the Net Present Value (NPV), and perform a sensitivity analysis on AMI alternatives. The model is intended to evaluate the economic feasibility of advanced metering options and provides the capability to select scenarios and compare results in real time.

Objectives of the Model The primary objective of the model’s development is to provide meaningful results while supporting the decision-making process regarding the systematic implementation of AMI. Jacobs used various scenarios and a range of assumptions to provide further confidence in the potential value of this approach for DWR and its customers by identifying and assessing each potential positive and negative quantifiable impact.

Methods used to meet the primary objective are as follows:

• Identify and assess each potential positive and negative impact of AMI

• Estimate the cost and benefit of each impact along with an appropriate phasing plan

• Calculate the NPV of the phasing plan to standardize costs in 2018 dollars

• Identify and document other non-quantifiable benefits to help further inform the decision

Since many of the exact values for inputs and assumptions will remain unknown until procurement is complete, the inputs are defined by a confidence range and the outputs of the model are presented as a set of ranges with minimum, maximum, and expected values serving as the reported figures. Minimum and maximum bounds are determined through a combination of reported information from DWR and from Jacobs’ extensive industry knowledge and experience. This approach provides an understanding of the sensitivities for each business case driver. During implementation, these outputs can also be used to improve risk management and provide focus on key value drivers.

Assumptions Assumptions have been incorporated into the model based upon input received from DWR staff and are presented below:

General Assumptions

Discount Rate 4.65%

General Escalation Rate 2.10%

Labor Escalation Rate 1.74%

Revenue

% Increase in New Connections 1.03%

% Annual Rate Increase (Volumetric) 0.55%

Definitions of assumption terms are as follows.

• Discount Rate – As provided by DWR, typically calculated as the weighted average cost of capital (WACC) and used in the NPV calculations

• General Escalation Rate – As provided by DWR, typically calculated as the inflation rate and used to increase future years’ non-labor costs

• Labor Escalation Percent – As provide by DWR, typically calculated as the inflation rate and used to increase future years’ labor related costs

• % Increase in New Connections – Assumed increase in the number of customers

• % Annual Rate Increase (Volumetric) – Anticipated annual increases in the volumetric water rate

Other financial inputs were developed in close collaboration with staff to establish appropriate cost estimates, time frames, and labor rates; these can be easily modified and updated in the Excel model as new information becomes available. The model also allows for adjustments to be made over time to reflect financial, performance, as well as other programmatic or strategic changes in the department.

Scenario Definitions

Base Case

The Base Case scenario assumes that DWR will replace meters as they reach one million gallons of total throughput (status quo). These quantities have been smoothed slightly to keep total yearly replacement quantities consistent with the historical replacement levels. The anticipated number of meters needed to be replaced over the next 20 years is shown below in Figure 2.

Figure 2: Graphical representation of the status-quo meter replacement schedule

5,000

10,000

15,000

20,000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Business Case Year

Base Case Meter Replacement

Jacobs developed an asset management plan for the existing assets that calculates the average yearly consumption for each meter in the fleet to determine when that meter would reach a million gallons of usage. A proxy of 66K gallons/year (5,500 gallons/month) was used as a proxy for meters with unknown install dates that prevented Jacobs from completing their actual usage trend analysis. This asset management plan, or Base Case, is what Jacobs has identified as the minimum requirement to continue to own and operate the current meter reading system.

The lifecycle cost of the asset management plan (Base Case) is approximately $24 million over 20 years.

AMI Scenarios

As a comparison to the Base Case, Jacobs has developed two AMI alternatives. These scenarios serve to illustrate the range of cost-benefit results (book ends) that would be available with AMI. The two alternatives are:

1. Fixed Network: a high-powered, low infrastructure system. Sensus pricing has been used as proxy.

2. Cellular Network: a cellular system, which does not require any network infrastructure. Badger pricing has been used as proxy.

Cost/Benefit Categories At the highest level, all categories are either considered a positive benefit or a negative impact. Positive benefits include additional revenue, cost savings, and avoided costs, while negative impacts include capital costs and additional operating costs. These categories are further divided into individual business case elements which are discussed in further detail below.

Positive Financial Benefits The positive benefits include the financial advantages of each scenario that were quantified and included in the net present value calculations. These benefits fall into one of three categories: additional revenue, cost savings, and avoided cost.

Additional Revenue

AR1: New Meters, Improved Registration

Minimum Expected Value Maximum

Base Case $ - $ - $ -

AMI $ 3,740,000 $ 7,020,000 $ 11,700,000

The first identified benefit relates to meter accuracy. Mechanical water meters have sensitive moving parts that wear out and start to slow down (under-register) over time. This “slowing” results in lost revenue because some water is flowing through the meter un-registered.

AWWA studies have identified that traditional mechanical meters can under-register as much as 6% over 20 years. That assumes, however, a national-average water quality and a mix of meter types and manufacturers. Because DWR has actual meter test data, Jacobs was able to use that data to calculate volumetric losses instead of using the AWWA standard. Note: DWR has not experienced 6% under-registration.

The meter test data provided by DWR included a variety of meter types; for this analysis, Jacobs only used residential meters as these are the only meters included in the improved registration calculations (larger meters are generally linked to commercial accounts which can vary much more in their usage patterns than residential and are therefore excluded). Weighted average accuracy for each meter was determined using a 15%, 50%, and 35% coefficient for low, medium, and high flows, respectively.

Anomalies were removed, and the remaining data points were grouped into 100K gallon cohorts. The average accuracy of each cohort was then plotted against measured accuracy.

Three trendlines were created via a simple linear regression to estimate the expected degradation over a million gallons, all of which appear to be significantly better than the AWWA standard. These three trendlines were then used as the minimum, maximum, and expected value for the improved registration calculations.

Figure 3: Graphical representation of DWR's meter test data and trendlines

• Minimum Trendline (Orange): A regression which assumes that the significant low points are potential anomalies and excludes them. Only meters with expected performance or better-than-expected performance are used to calculate the regression (equivalent to 0.8% under-registration at a million gallons of usage)

• Expected Value Trendline (Green): A best-fit regression that examines all data points and includes all age cohorts in the regression (equivalent to 1.5% under-registration at a million gallons of usage)

• Maximum Trendline (Purple): A regression which assumes that the significant high points are potential anomalies and excludes them. Only meters with expected performance or worse-than-expected performance are used to calculate the regression (equivalent to 2.5% under-registration over a million gallons of usage)

The benefit of improved registration is based on eliminating the under-registration of the current meter stock and a function of the systematic replacement of all aged meters within 3-4 years. The benefit is quantified as those meters that are replaced before they would have been replaced in the asset management plan (Base Case).

Calculated volumetric loss, based on specific meter populations being replaced “early” (by estimated time to reach one million gallons of throughput) was then multiplied by the average residential price of water/sewer in each year, to produce a dollar-value impact of under-registration. This results in approximately $7 million in additional revenue over 20 years.

Note: Demand growth and planned revenue increases were included in these calculations.

90%

92%

94%

96%

98%

100%

102%

0 5 10 15 20 25

Weighted Average Accuracy for Blocks of 100,000 Gallons

2018 Degredation Curves based on Meter Test Data

Test Data Min 99.2% @ 1M EV 98.5% @ 1M Max 97.5% @ 1M

Cost Savings

CS1: Reduced/Eliminated Bermex Contract

Base Case $ - $ - $ -

AMI $ 29,800,000 $ 30,910,000 $ 32,020,000

DWR currently expends approximately $2.7 million a year on a Bermex contract to perform all reading activities as well as move-in/move-out workorders, disconnects, reconnects, and several other ad-hoc operational tasks. Approximately 90% of the existing contract is for reading services. With an AMI system, all these activities will no longer be required as reads are automatically acquired, transmitted, and billed over the field network and processing systems. Even re-checks can be accomplished from a computer in the office. And many other activities, while not eliminated, will be significantly reduced.

Under an AMI scenario, DWR will have two options: renegotiate a significantly smaller contract with Bermex to continue performing any remaining activities or eliminate the contract entirely and hire in-house staff. The table below contains the unit pricing and quantity for each activity in the 2018 Bermex contract along with estimates compiled by Jacobs of how these activities will change after an AMI system is installed.

Activity Quantity Unit Cost % Retention Hours/Activity

General Reading Services 3,060,000 $ 0.71 10% 0.014*

Vaulted Reading/Entry Required Reading Services 16,800 $ 3.00 0% 0.014*

Large Remote Meters Reading Service 6,000 $ 2.00 0% 0.014*

Detector Check 3,300 $ 2.00 0% 0.25

Move-In Orders 12,000 $ 5.00 25% 0.25

Transfer Orders 25,500 $ 5.00 25% 0.25

Move-Out Orders 13,000 $ 5.00 25% 0.25

Lock-Off Orders 20,000 $ 5.00 100% 0.25

Reconnect Orders 20,000 $ 5.00 100% 0.25

Meter Lid Replacement 1,250 $ 3.00 25% 0.25 *Based on 425 reads/person/day

Figure 4: Bermex contract pricing and cost savings assumptions

The 10% retention under ‘General Reading Services’ is to perform an annual inspection of each meter which was a task requested by DWR to ensure meter box condition and location are monitored and maintained. Move-ins, transfers, and move-outs will still occur, but the majority of them can be performed by simply placing a ‘flag’ on the account to indicate that there should be no usage after the customer has vacated. A truck roll would only be necessary if unauthorized usage (automatically detected by the system) occurs.

How Bermex will respond to such a reduced contract cannot be determined at this time. However, based on Jacobs’ analysis, if Bermex decides to raise its unit prices by more than 75% for the remaining activities, it will be cheaper for DWR to hire in-house staff. The retained level of effort equates to approximately 13 new full-time employees. Jacobs has assumed a fully loaded TA3 salary ($68K/year) and an additional vehicle for each new hire. The total cost of these 13 new employees and their associated vehicles is just under $1 million a year – nearly $1.7 million less than the yearly Bermex contract. It is worth noting that these hires would not all need to be brought onboard all at once and could be obtained at a rate of 3-4 a year over the course of the 4-year implementation schedule. It is also worth noting that these additional employees would only be necessary if DWR continues the same activities currently being completed by Bermex resources; changes in meter check policies or operational optimization could have an impact on the number of FTEs needed.

The decision to retain Bermex is still highly dependent upon how Bermex responds to a reduced level of effort in the contract. Should they maintain their current unit pricing it will be significantly less expensive to extend the contract than hiring additional staff in-house.

Avoided Costs AC1: Existing Meter Replacement

AMI vs Base Case $ 21,660,000 $ 24,070,000 $ 26,480,000

By systematically replacing the bulk of DWR’s meter stock (meters with less than 200K gallons of total throughput will be retained) with an AMI system, DWR would be choosing to replace the existing meters all at once, instead of continuing to repair and replace them as they reach a million gallons of usage.

The new AMI system will eliminate the need to continually replace meters over the next 20 years. This is a good example of an avoided cost.

Jacobs has developed a budgetary cost estimate for the Base Case. As mentioned above, the Base Case assumes that DWR will continue to replace aging meters with new ones, year after year, as they reach a million gallons of throughput. Replacement begins with the cache of meters that have already exceeded a million gallons (approximately 29K meters) and proceeds in the following year with the next group of meters anticipated to reach a million gallons of usage, and so on. The Base Case further assumes that, year-after-year, meter replacements will be performed by the Meter Shop staff while also performing their regular duties of maintaining the remainder of the system. The avoided cost of replacing the system this way is approximately $24 million over 20 years.

Note: Jacobs has used the current contracted meter pricing from Badger for these calculations. Going through a competitive procurement process will likely have the added benefit of securing further reductions in unit costs from Badger and others. The expected pricing from Badger in an open market RFP is used to calculate the budgetary cost in the Cellular AMI Scenario in the Negative Financial Impacts section.

AMI Alternatives

Both AMI scenarios presented in this study represent full system-wide deployments of AMI, which is required to achieve economies of scale for material and installation pricing, to deliver uniform benefits across the customer base, and to effectively realize benefits such as reduced meter reading costs.

We have not proposed an alternative to implement AMI only for non-residential customers, since this would have limited benefit. Although the non-residential customers represent nearly 50% of water sales, they do not represent 50% of the AMI benefits. DWR would not realize additional revenue from improved registration because the larger meters are not typically replaced (they can be calibrated).

DWR would not realize the full cost savings from reduced meter reading since all of the residential meters would still need to be read. And DWR would not realize avoided costs of meter replacement since larger meters are not part of the meter replacement program. If preferred, DWR could certainly prioritize the non-residential accounts to be upgrade first, as part of a broader AMI upgrade program, but without the residential accounts, the overall benefits are likely to be nominal.

Negative Financial Impacts This section includes the budgetary cost estimates that Jacobs has developed for the initial capital cost as well as the ongoing operation and maintenance requirements for each AMI scenario. These costs are grouped into the following three categories: Capital Cost, Additional Operating Cost, and Reduced Revenue.

Capital Costs

The initial capital cost includes the cost to replace meters and install radios, repair or replace boxes, add fixed network infrastructure, and pay for project/program management and integration services. Both of the AMI scenarios are similar, although each has their own specific features.

Note: These scenarios are not intended to provide the full range of potential AMI solutions, rather they are offered as bookends for potential system costs.

• The fixed network scenario will require a build out of network infrastructure that will be maintained by DWR or a third-party firm as part of a yearly maintenance fee.

• The cellular network scenario will not require any network infrastructure but incurs a monthly fee for each connected customer/endpoint in the field.

Both scenarios are expected to take approximately 36 months to implement and are assumed to be paid for over four consecutive fiscal years. The network (if needed) will be deployed during the first 6 months of the project (first fiscal year), with 1/6 of the meters being installed every six months thereafter (1/3 during second and third fiscal years, and 1/6 in the fourth fiscal year).

In both scenarios, it is assumed that most lids will need to be replaced, including lids made from a custom mold for the old oval boxes. Based on initial estimates from DWR staff, we have assumed approximately 217,700 new lids will be acquired at the same frequency and timing of meter replacements. This results in a cost of approximately $4.4 million.

Jacobs has also developed a budgetary cost estimate for project and program management ($4.2M) and included a buffer of contingency ($500,000) for unforeseen details related to exact scope of services, quantities, and field conditions. Including the billing system interface, training, postage, and warehosing fee and the total professional services cost is approximately $5.5 million, which is applicable across all AMI scenarios.

Note: Ongoing O&M costs have been developed for the expected 20-year life of the system and discounted back to 2018 dollars using an NPV. The initial capital cost, although spread over four fiscal years, has not been discounted, because the initial contract value will be in today’s dollars.

Fixed Network Scenario

Meters $ 23,620,000 $ 26,240,000 $ 28,870,000 Boxes & Lids $ 3,960,000 $ 4,400,000 $ 4,840,000

Radios & Network $ 14,850,000 $ 16,500,000 $ 18,150,000 PM & Integration $ 4,970,000 $ 5,520,000 $ 6,070,000

Total $ 47,400,000 $ 52,660,000 $ 57,930,000

Jacobs has assumed that a high-powered fixed network vendor will require a total of twenty-five collectors for the network infrastructure; we also ran a sensitivity analysis with 100 collectors in the model, which only resulted in a reduction of $1.8 million in NPV and one additional year of payback (inclusive of additional O&M for backhaul and maintenance). This demonstrates that the cost-benefit results are not highly sensitive to the network infrastructure, since it represents such a small portion of the overall cost.

We have also included a mobile collector to provide backup reads, as well as three handheld computers with belt clip transmitters for maintenance of endpoints.

Should DWR pursue a fixed network solution, it could be possible to partner with Georgia Power (or other municipality) through a competitive process, which could produce some political and perception benefits, however longer-term system surety and ongoing cost sharing arrangements need to be addressed to mitigate risk exposure for DWR.

Cellular Network Scenario

Because Badger uses cellular connectivity for its AMI system, DWR will not be required to purchase, install, or maintain any fixed network infrastructure. The analysis includes a mobile collector to provide backup reads, as well as three handheld computers with belt clip transmitters for maintenance of endpoints.

Including the costs of lids, program management, and contingency (previously discussed) the total expected initial capital cost for this scenario is $51.6 million.

Meters $ 20,650,000 $ 22,940,000 $ 25,240,000 Boxes & Lids $ 3,960,000 $ 4,400,000 $ 4,840,000

Radios & Network $ 17,560,000 $ 19,510,000 $ 21,460,000 PM & Integration $ 4,060,000 $ 4,510,000 $ 4,960,000

Total $ 46,230,000 $ 51,630 000 $ 56,500,000

Additional Operating Costs

Network Fees and Maintenance

Jacobs has estimated the operating costs of each scenario for the expected 20-year life of the system.

These costs include the hosting fee, software licensing fees, collector/radio backhaul, the collector warranty fees, and collector maintenance contract.

Again, both the AMI scenarios are similar, although each has their unique aspects:

• Both scenarios: we have included a hosted meter data management system, which will provide data storage for hourly or 15-minute interval data.

• Both scenarios: we have included annual software licenses for the meter data management system, the utility portal (to check the health of the system), and a vendor provided customer portal, so customers can see their usage patterns. Note: the cost of WaterSmart or other 3rd party customer portals have not been included.

• Fixed Network scenario: we have included the cost of network backhaul for the new collectors.

We have also included the cost of an extended warranty for the collectors as well as an on-call maintenance contract to cover the labor for collector maintenance and replacement.

• Cellular Network scenario: we have included the cost of cellular backhaul for each individual radio.

The annual O&M cost for each of the above items is totaled over 20 years and discounted back to 2018 dollars for comparison. A breakdown of the expected values for each vendor is provided below.

Minimum and maximum bounds are ± 10% for each scenario.

Fixed Network Cellular Network

Hosting Fees $ 180,000 $ 180,000

Software Licensing $ 180,000 $ 207,000

Backhaul $ 7,500 $ -

Collector Maintenance $ 32,000 $ -

Collector Electrical Usage $ 2,600 $ -

3rd Party Backhaul $ - $ 841,000

Yearly Total $ 403,400 $ 1,228,000

20-Year NPV $ 5,420,000 $ 14,070,000

New AMI Labor

DWR currently spends approximately 5,200 hours a year on assessing re-reads/hi-lows and ad-hoc meter repairs. Under both AMI scenarios, this work will be eliminated as reading issues can be conducted remotely and the current meter fleet will no longer be in place. Meter repairs, however, will still be necessary on the new AMI system as endpoints and meters fail for a variety of reasons. The industry has begun to include a failure rate “ceiling” in its contracts that is typically 2% for device failures and 1.5% for read success. Assuming 45 minutes to address either of these type of workorders, this means that at a maximum, DWR can expect 4,800 new hours of work maintaining the new system. This represents a difference of 400 hours from existing work, indicating that no new staff will be required for

AMI-specific maintenance.

Economic Evaluation Each cost and associated benefit of AMI are calculated separately and added together for each year in the 20-year project lifecycle. The result of adding all benefits and costs together is a series of cash flows (See Figure 5). The cash flows are negative in the first few years, due to the initial capital cost of the project. The project then begins to generate positive cash flows and accrue financial benefits for DWR.

Figure 5: Expected Value Cash Flow for Fixed Network Alternative

Model Results Jacobs performed several sets of analyses, with the expected values for each scenario presented below.

The assumptions and methodologies are described further throughout this document.

The terms are defined as follows:

• Initial capital cost is the budgetary cost to acquire and install the system

• Average annualized benefit is the quantified benefits each year minus cost each year – this should be a positive number

-$50

-$40

-$30

-$20

-$10

$0

$10

$20

$30

$40

$50

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

M il li o n s

Expected Value Cash Flow over 20 Years

Annual Cost Annual Benefit Cumulative Net

• Simple payback is how long it takes for sum of benefits to exceed the sum of costs – this should be less than the expected life of the system, which is 20 years

• NPV is the total sum of costs and benefits, discounted back to today’s dollars – this should be a positive number

• ROI is the average annualized benefit represented as a percentage of the initial capital cost – this should be greater than the weighted average cost of capital, which is 4.5%

AMI Business Case Outputs: Fixed Network Cellular Base Case

Initial Capital Cost $ 52,660,000 $ 51,360,000 $ -

Average Annualized Benefit $ 4,490,000 $ 3,795,000 $ -

Simple Payback (years) 12 14 0

Net Present Value $ 8,500,000 $ 830,000 $ (24,070,000)

Return on Investment (ROI) 9.0% 7.0% 0.0%

Both AMI scenarios share a similar order of magnitude for the initial capital costs. However, because of the relatively high cellular data fees, the average annual benefit (and total NPV) is slightly lower than the fixed network scenario. That said, the payback in both scenarios is within our recommended duration of 15 years (3/4 of the expected life of the system), the NPV is positive (benefits exceed the cost), and the ROI is higher than the cost of money. This result indicates a strong financial justification for the project.

Other Benefits and Opportunities Several additional benefits may be available to DWR through the system-wide implementation of AMI, which have not been quantified or included in the business case model. The two main reasons for not including these benefits are: 1) the benefit is not easily determined, or 2) the benefit cannot be directly linked to the installation of an AMI system. For some utilities, however, these non-quantified benefits are just as, or even more, meaningful than the financial drivers.

Improved Customer Service - AMI enables a richer customer service experience, which is difficult to quantify, but it is a key benefit of AMI. Not only will it enhance DWR’s relationship with its customers, but AMI will also improve the job satisfaction of customer service representatives by providing them with the tools and data to do their job more effectively and answer customer's questions. Some of the benefits relating to improved customer service are described below. These benefits have not been quantified in the business case calculations and simply exist as qualitative support for system-wide AMI.

• Web/Mobile Platforms – The availability of hourly data gathered from AMI can be presented to customers on their bill as well as via web/mobile platforms. Customers will be able to monitor their water usage patterns over time. To have this data available on-demand may not actually result in a financial benefit for DWR, though it is appreciated by customers and instills confidence in bills through transparency. Over time, it can serve to reduce the total volume of calls to the call center, as customers may start to answer simple consumption related questions without direct involvement of a utility representative. Customer self-service via the web portal will allow for customers to set leak detection and other types of notifications and make downloadable consumption data available.

• Choice of Bill Payment Date to Customer – With traditional manual meter reading, there is a cost savings if all meters in a neighborhood can be read on the same day. However, with AMI there is no cost advantage to reading meters in the same neighborhood on the same day.

Meters can be read on any day in any location at no marginal cost. Therefore, DWR could allow customers to choose an alternative billing schedule with a due date that better fits their personal finances or business processes. This could be of value to customers on fixed incomes.

• Consolidated Billing for Customers with Multiple Meters – In some utilities, customers have multiple meters in different locations throughout the service area. With AMI, billing can be consolidated for these customers by syncing billing periods across all accounts belonging to a single customer. The advantage is the customer will have one due date to follow for all accounts, and all service locations will have the same service period for easy comparison between sites. An example might include all public schools within the County.

• Elimination of Meter Reader Access Issues – This is an important aspect for the safety of meter readers, but also helpful for commercial and industrial customers that are located behind locked compounds and want to maintain their site security.

• Improved Billing Service – Interval data will create customer service gains from: reduced or eliminated meter read estimates, faster turnaround time related to off cycle reads and final reads for move-in/move-outs, and reduce delays between read date and bill date, which will strengthen the price signal on customer consumption.

Environmental Benefits AMI will also reduce DWR’s environmental impact through several potential benefits, which are difficult to quantify, but nonetheless real. Many of these will accrue without specific effort from DWR staff.

• Reduced Greenhouse Gases – Reduced carbon emissions and greenhouse gases will be possible because of fewer miles driven by meter reading vehicles. In addition, there will be less fossil fuels purchased, because of less vehicle usage.

• Water Conservation and Water Efficiency– Water is a limited resource that should be protected and managed carefully, and AMI is one of the few ways that utilities are able to provide actual usage feedback to customers and enable them to become informed stewards of this critical resource. This is considered a soft benefit because, in general, reduced water usage is an overarching goal of most water utilities and considered “the right thing to do.” Supporting conservation improves the utility’s reputation with various stakeholder groups and environmental concerns. Reduced consumption can also be an offset against the need to secure future sources of supply.

• Prevent Water Contamination – Backflow detection (alerts of reverse flow through the AMI utility management portal) can be used to alert cross-connection and maintenance teams to help mitigate risks to the distribution system.

• Improved System Planning – Availability of peak day/peak hour water usage may enable DWR to reduce energy use for pumping and/or shift some demands off peak to defer the need to increase supply capacity in the distribution network.

Operational Benefits Another area of non-quantified benefits is those that impact operations. While some of these are somewhat challenging to measure and indirect, there have been case studies of other utilities realizing value associated with AMI across these categories.

• Improved Drought Management – AMI would be helpful in times of water restrictions, when residential customers are assigned specific allotments and hefty penalties for exceeding them.

In such times, AMI will be able to alert customers as they approach their limit and help administrative staff be more efficient at implementing rationing and addressing excessive use penalties.

• Improved Health and Safety – Meter reading is a high risk job (repetitive movement and physical threats from people, animals, and cars) but by reducing field activity, AMI can result in less risk of injury for meter readers.

• Hydraulic Model Calibration – Hourly interval data could be used to establish diurnal patterns for various customer types to help DWR improve the accuracy of their hydraulic model.

• Improved Cost of Service Analysis for Ratemaking – Hourly interval data could help DWR better understand the impacts that different customer groups have on the system, improving the cost of service analysis and enabling rate features such as peak use charges.

• Right-Sizing Large Meters – Interval data may allow DWR to monitor flow rates for each large meter to determine if the meter is correctly sized for minimum and maximum flows. This analysis could be incorporated as a standard meter flow rate analysis report that would flag or identify “over-sized” meters.

Conclusion The cost-benefit analysis, as described in this report, provides strong financial justification for deploying AMI system-wide. Together, the existing replacement schedule and Bermex contract total $55 million over the next 20 years (discounted back to today’s dollars) while the cost to deploy an AMI system is less than $53 million. The investment appears to make sense from that simple cost comparison, however when adding the increased revenue from newer meters and all the customer service benefits that will result from having access to real-time usage, the project is well justified.

The overall results (based on assumptions and estimates developed in this BCE) of AMI compared with the Base Case indicate that a full AMI deployment would provide DWR with an average annual benefit of approximately $4.5 million, a payback of 12 years, and a return on investment of 9%. Both the fixed network and cellular scenarios covered in this analysis render a positive net present value.

Due to the positive outcomes of this comprehensive analysis, it is recommended that DWR pursue a systematic deployment of AMI. It is acknowledged, however, that capturing this value is not immediate nor guaranteed without careful planning. Success is dependent upon identifying, designing the program around, and measuring the progress toward key outcomes.

Path Forward As DWR seeks to take steps toward designing its overall AMI program, Jacobs recommends a 5-step roadmap for its AMI program, as shown in the diagram below. The first step has been completed.

Figure 6: AMI Program Roadmap

Define Key Success Criteria and Measurement Strategy There are some fundamental processes to be followed in an AMI implementation program, but there are also customer, organizational, and stakeholder related choices DWR will need to make, based on its desired outcomes for the program. Once these objectives are set, DWR will be able to measure the progress and success of its program, and confidently report its results to customers, staff, and other stakeholders in ways they can appreciate. Recommended next steps, to be completed over one to two months, include the following:

• Identify the key success criteria that DWR wishes to realize through AMI, including customer and stakeholder satisfaction metrics, internal drivers and operational efficiencies, and financial goals;

• Establish the means to measure progress toward the identified key success criteria, as well as establishing the current baseline;

• Develop a communication plan for both internal and external stakeholders based on the key success factors;

• Devise a high-level action plan to include external stakeholder outreach activities, skills training, internal communication, changes in position responsibilities, business process changes, and system integration that sets out the steps DWR should take to ensure its AMI program delivers on expectations.

Proceed to Procurement Jacobs would like to continue helping DWR as the owner’s engineer to provide expert advice and oversee the procurement and implementation efforts for AMI. These are once-in-a-career projects and there is very little opportunity for on-the-job training. The next two phases of the project are RFP Development and Vendor Evaluation, which will take approximately 6 months and will require that the Purchasing Department is engaged early. Tasks involved in that process are summarized as follows:

• RFP Development – The RFP should be open to all meter vendors to encourage competition and specifications that exclude any one-type of system should not be used. The evaluation process will shortlist the proposals to ones that meet the needs of DWR. Existing meter population (quantity and sizes), customer addresses, and specific locations/height of DWR facilities will need to be provided as well as standard contract language.

• Vendor Evaluation - Our recommended approach is to request both technical and cost proposals initially, score them to determine a short list, request short list vendors present to DWR, and then request best and final cost. We also recommend using 50% weighting for technical response (4 criteria), 30% for the presentation, and 20% for cost. This approach has been developed to provide transparency to DWR during the process and ensure the best possible price for the selected AMI system.

These recommended steps will allow DWR to bridge the gap between the business case results and developing a path to implementation: defined success criteria, engagement strategies to enlist support from stakeholders, a method to measure results in meaningful ways, and a clear path to selecting an AMI vendor.

Appendix A: Glossary of Terms A list of technical terms and their definitions are provided below:

• AMI – Advanced Metering Infrastructure (also referred to as fixed-network)

• AMR – Automated Meter Reading (also referred to as drive by)

• Collector – A receiver of data transmitted by endpoints. Can be referred to generally as

“network devices”. Collectors are generally placed on existing infrastructure (i.e. city hall, water tower, telephone pole, etc.).

• Drive-by read – A meter read obtained through a mobile collector, typically used while driving a vehicle through routes.

• Endpoint –…