- Email: [email protected]
benefit analysis of an AMR system to reduce electricity theft and maximize revenues for Électricité du Liban
Applied Energy 76 (2003) 25–37 www.elsevier.com/locate/apenergy
Cost/benefit analysis of an AMR system to reduce electricity theft and maximize revenues for E´lectricite´ du Liban Raymond F. Ghajar*, Joseph Khalife Lebanese American University, PO Box 36, Byblos, Lebanon Accepted 22 January 2003
Abstract One of the largest pitfalls for any distribution network is the level of energy losses suffered by the system. These losses fall into two categories: technical and non-technical. Technical losses depend largely on the physical properties of the network, while non-technical losses (sometimes a more significant form of losses) are the result of theft or fraud caused by meter tampering, false reading, illegal connections or unpaid bills. In Lebanon, the levels of total losses are around 50% resulting in an annual deficit of more than 225 million US dollars. Despite the frequent breakdowns of the system and evidently unsustainable financial-losses, political consideration makes the sustained pursuit of electricity thieves low on the list of priorities. To overcome these hurdles, the national electricity company in Lebanon, E´lectricite´ du Liban (EDL), studied the possibility of using automatic meter-reading (AMR) technology to modernize electricity metering, billing and collection, minimize fraud and maximize revenues. The results of this study and a cost/benefit analysis of the proposed system are summarized in this paper. # 2003 Elsevier Ltd. All rights reserved. Keywords: Automatic meter-reading; Energy economics and management; Case study
1. Introduction The distribution of electricity has long been seen as a safe business venture. With secured earnings and a captive market, there appears to be little that could cast a * Corresponding author. Tel.: +961-9-547254; fax: +961-9-547256. E-mail addresses: [email protected] (R. F. Ghajar), [email protected] (J. Khalife). 0306-2619/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0306-2619(03)00044-8
26
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
shadow over such an investment. However, despite the apparently rosy glow of distribution network finances, the spectre of electricity theft lurks in the shadows of many developing countries, seriously impinging on their fundamental economics [1]. One of the largest pitfalls for any distribution network is the level of energy losses suffered by the system. Distribution losses depend largely on the physical properties of the network, such as conductivity and resistance of the cables, the number of voltage transformations made and the size of the network. The second, and sometimes more significant, form of losses comes under the heading of non-technical losses (i.e. theft and fraud). Several types of non-technical losses exist:
Fraud (meter tampering, false reading, etc.) Theft (illegal network connections) Collection (failure to collect monies owed) Billing (companies do not know how much electricity has been consumed).
All of these factors are exacerbated by the fact that detection of theft from a distribution network is extremely hard to identify. Essentially the only way to determine the level of losses, both technical and non-technical, is to subtract the level of payment for electricity delivered across the network from the value of power that was generated, the net figure thus produced gives an indication of the total losses. Attempts to determine empirical values for theft are, more often than not, simply estimates based on levels of detection. As Lebanon was being rebuilt after 20 years of civil war, the national powercompany, EDL, started to progressively modernize its infrastructures to meet the growing demand of residential, commercial and industrial customers [2–4]. With new generation and transmission facilities already built, EDL is currently attempting to improve its electricity metering; billing and revenue collection in order to increase revenues and reduce fraud, which accounts for almost 50% of missed revenues today [5]. The objective of this paper is to present a cost/benefit analysis of the proposed system for modernizing the electricity metering, billing and collection processes for EDL customers. Currently, these customers are being metered using a variety of old and new electromechanical meters. The metering process is manual, prone to errors and does not provide EDL with reliable methods to readily check bills, identify problematic customers, and reduce fraud. Several types of meter-reading systems were investigated for their applicability to Lebanon, including mobile, fixed and power-line carrier systems. The proposed system will ensure that EDL’s metering activities would become as advanced as those of modern utilities, reduce fraud and have a reasonable pay-back period. Most importantly, with the proposed system, EDL will regain the confidence of its customers, and would become a highly reputable and revenue-generating organization. The results presented in this paper will show that the proposed modernization project is technically and economically feasible.
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
27
2. Electric-metering solutions Electric and water utilities around the world are progressively moving towards automatic meter reading (AMR) technology to provide value-added services to their customers [6–8]. Most AMR deployments involve mobile AMR technology, where the meter reader uses handheld computers to communicate with the meter from a short distance. Drive-by solutions are also popular, where the meter-reading computer is installed in a van and the meter reader does not have to walk. The benefits of a mobile AMR for electricity customers are: 1. Faster reading of the meter, allowing for the reduction of the number of meter readers and reduction of operating costs. 2. Better accuracy of reading, allowing for the reduction of bill complaints. 3. Fewer contact with adverse elements (dogs, inclement weather, etc.) for driveby solutions. The additional benefits of a fixed AMR for the same customers are: 1. More frequent reading, for customers whose daily consumption is important. 2. More complex billing rates, allowing for incentives and marketing actions to retain and gain more customers where competition exists at the retail level. 3. Total elimination of the meter reader, in particular for difficult areas (high crime, very remote, hard to access, etc.). 4. Possible control of electric demand and the ability to perform on-request meter reading. The system recommended to EDL is a mobile automatic meter reading (AMR) system with radio frequency (RF) communication because it is the least expensive, simple to install, quickest to implement, easiest to upgrade, compatible with other AMR systems and field proven [5]. The equipment that is needed for the automatic metering system includes meters, communication systems, network-management software as well as data collection and storage software.
3. Technological alternatives These rely on a mobile automated meter reading system, which centralizes the meter reading and data collection at EDL’s headquarters and interfaces with the current billing system. This solution will give a proper incremental modernization that is compatible with EDL’s current financial and human resources. The proposed AMR system will include the following components: 1. Handheld computers that will be used in high-density areas for entering meter data and acknowledging payment of bills. Each unit communicates with the central server using a data network via a personal computer (PC) and a
28
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
2.
3.
4.
5.
6.
modem. The unit can communicate with the meter using short-distance radio and it also includes a keyboard for manual data-entry. Vehicle-based units will be used in low-density areas for reading meter data and acknowledging payment of bills. Each unit communicates with the central server using data network via a modem. The unit can communicate with the meter using a short-distance radio; it can also be used manually. Metering and route-management software, which will guide and control the meter reader data entry. The software will be installed in the handheld units and vehicle-based units. Collection-management software, which will control the collection route and the data entry for acknowledgment of bill payments. The software will be installed in the handheld units and vehicle-based units. Central management software installed at the central server, which stores the meter and collection data and interfaces with the current billing system. This interface will allow for daily transactions with the billing system, including updates of customer accounts, billing process, etc. New meter equipment, depending on the types of customers: For low-consumption customers (i.e. residential customers), the meters will be left unchanged. Data will be manually entered in the handheld units or vehicle-based units. For high-consumption customers (i.e. mostly commercial customers at the low-voltage level), the meters will be replaced with new ones (electromechanical or electronic) equipped with short-distance radio communication modules. The handheld units and vehicle-based units will communicate with the meters remotely, thereby eliminating all physical contact with the meter and reducing reading errors to a minimum. For problematic customers (i.e. residents who are tampering with the meters, pirating electricity, etc.), an electronic meter will be installed and EDL should consider enforcing the existing regulations to better deal with such customers.
4. Projected costs A summary of the capital and operating cost requirements for 943,000 customers (estimated number of EDL customers), for Alternative 1, electromechanical (EM) meters equipped with RF communication modules, and Alternative 2, electronic meters equipped with RF communication modules are summarized in Table 1. These cost estimates are based on budgetary-cost information gathered from suppliers of metering and billing systems. The capital requirement projections for both alternatives (electromechanical and electronic) are presented in Table 2. These projections are based on purchasing all 135 handheld computers, the regional host computers, and associated data collection software during the first year of implementation. Additionally, the
29
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37 Table 1 Budgetary cost estimates for the proposed AMR solutions Capital costs
Cost ($/item) Electronic
A. Component costs Meter w/RF Handheld computers w/RF Regional host computer Software: reading/scheduling/transfer
200
Quantity
Total
EM
Electronic
EM
86 943,000 4900 135 2000 12 200,000 1
$188,600,000 $661,500 $24,000 $200,000 $189,485,500
$81,098,000 $661,500 $24,000 $200,000 $81,983,500
Total
Total
B. Training costs Cost ($/h) Meter readers Supervisors/schedulers
4 4
Quantity 2160 480
$8,600 $1,900 $10,500
$8,600 $1,900 $10,500
C. System setup costs Cost ($/h) Data entry Network engineer Data analyst Programmer Meter installation
5 9 9 9 4
Quantity
Total
Total
1044 348 348 348 471,500
$5,200 $3,100 $3,100 $3,100 $1,886,000 $1,900,500 Total capital costs $191,396,500
$5,200 $3,100 $3,100 $3,100 $1,886,000 $1,900,500 $83,894,500
Cost ($/h)
Total
D. Optional components Cost ($/item) Billing system Remote disconnect Drive-by metering Advanced drive-by metering Spot billing Annual operating costs A. Operating costs Network Engineer Data Analyst Programmer
1,200,000 25 15,000 130,000 3800
9 9 9
Quantity 2088 2088 2088
$19,000 $19,000 $19,000 $57,000
B. Maintenance costs % Capital Cost Meters Handheld computers Host computers Billing system
0 10 10 0
Capital Cost 81,098,000 661,500 24,000 1,200,000
Total $0 $66,150 $2,400 $0 $68,550
30
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
Table 2 Capital requirements for the proposed AMR solutions Meters
Year Year Year Year Year
1 2 3 4 5
Handheld Host System computers computer set-up systems
EM
Electronic
$0 $20,274,500 $20,274,500 $20,274,500 $20,274,500
$0 $661,500 $47,150,000 $0 $47,150,000 $0 $47,150,000 $0 $47,150,000 $0
$234,500 $0 $0 $0 $0
$14,500 $471,500 $471,500 $471,500 $471,500
Operating & Annual capital requirement maintenance
$125,550 $125,550 $125,550 $125,550 $125,550
Total annual capital requirements
EM
Electronic
$1,036,050 $20,871,550 $20,871,550 $20,871,550 $20,871,550
$1,036,050 $47,747,050 $47,747,050 $47,747,050 $47,747,050
$84,522,550
$192,024,250
annual operating-cost for maintenance of the computers, and the fully-loaded cost for three information technology professionals to support the data collection system are included in the first-year capital requirements. In years two through to five, the capital and installation costs of 236,000 meters per year (i.e. 943,000/4) are included in the capital requirements projection, plus the annual operating cost. In addition to alternatives 1 and 2, a hybrid solution (the recommended alternative) that includes the following components is also proposed: New meters equipped with RF communication modules (943,000; 86% electro-mechanical and 14% electronic meters). Handheld computers capable of reading up to 2000 meters per day (135 units). Regional host computers (12 units). Metering and route management software. Portable meter reading computer capable of reading up to 4000 meters per day with spread spectrum (optional). Vehicle meter reading computer capable of reading up to 25,000 meters per day with spread spectrum (optional). The cost figures for the hybrid solution are based on the pro-rated costs of the previous alternatives. The recommended alternative gives the electric utility the freedom to choose the appropriate metering technology for each customer depending on his status (e.g. problematic, large consumption, regular, etc.). In addition to the proposed technology alternatives, a number of new processes were proposed to provide EDL with better control of its current situation and properly position the utility for the transition into the new automated system. Since these processes have an impact on the benefit analysis described later in this paper, they are briefly described below:
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
31
Separation of meter-reading and bill-collection processes. Use of handheld computers for meter reading and validation of customer information. Enhanced classification and identification of customers. Placing responsibility on the customer to pay the bill and not on EDL to collect. Inclusion of separate receipt with each bill. Usage of electronic meters to verify suspected problematic-customers. Imposing strict penalties on problematic customers. Disconnection of non-paying customers. Set-up of regional collection centres or coordination with banks as collection centres. Providing incentives for customers to pay their bills with automatic bank drafts.
5. Schedule for phased implementation A project plan for the phased implementation of the modernized system is presented in this section. The recommended modernization effort consists of six activities: Step 1. Purchase and introduce hand-held computers in year 1. Step 2. Modify the billing and collection processes, revise the customers’ database, and conduct an advertising campaign to inform customers with procedures for bill payment and penalties in year 1. Step 3. Install new meters with RF communication modules in years 2 through 5, targeting high consumption and problematic customers first. Step 4. Introduce one vehicle-based meter reader in year 2 to free some of the meter readers to perform other activities (optional). Step 5. Begin executing the recommended procedures for dealing with problematic customers in Year 2. Step 6. Upgrade the billing system in year 4 or 5 to meet EDL’s growing requirements (optional). The schedule for this complete modernization effort as illustrated in Fig. 1 can be completed in 5 years.
6. Projected benefits The benefits to EDL of modernizing its electricity metering, billing and collection processes are described in this section. The benefits are identified in terms of the steps identified in the phased-implementation schedule. While it is possible to
32
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
Fig. 1. Five-year plan for modernizing the metering, billing and collection systems by EDL.
accurately estimate the costs for implementing an AMR system, placing a dollar value on the potential benefits is far more difficult. The following discussion of potential benefits attempts to identify the nature of the benefits and, where possible, provides some measure of those benefits in terms of dollars, either saved or collected. 6.1. Introduction of hand-held computers This first step involves the issuing of handheld (computer) units to EDL meter readers to replace the manual recording of meter readings. It also requires that a system of regional computers be deployed for communicating between the handheld computers and the existing billing software that resides in EDL’s central mainframecomputer. Existing meters are not replaced at this point and the following three basic benefits are noted: Fewer customer complaints and more timely payments are expected. Using handheld computers that communicate directly to the billing program should reduce the number of data entry errors that find their way into the billing system. This should improve customer confidence of the resulting billings and lead to more timely collections. Fewer customer complaints of billing anomalies should result because of the more accurate readings obtained with the hand-held computers. Holding meter readers more accountable will reduce fraudulent reading practices. The high/low value checking feature of the hand-held units and the promise of an appropriate monetary penalty should thwart these practices. The projected benefits of employing handheld computers combined with penalizing the problematic customers should amount to approximately $3.5 million in years
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
33
2, 3, 4, and 5 of the program [5]. Considering a start-up delay, it is expected that about half of that amount could be realized in the first year of the program. 6.2. Change collection system Under the current system, an EDL employee delivers the bill personally to the customer and collects payment. This system places the responsibility of receiving payment on EDL. It often requires multiple trips to the same customer and results in a slow, time-consuming process. This responsibility needs to be reversed, placing the responsibility on customers to pay their bills in a timely fashion. A number of changes were recommended in bill processing including changes in bill delivery and revenue collection. These changes should yield the following benefits. Quicker collections of the metered portion of the consumed energy. Elimination of the need for many, if not all, contract collectors and freeing up some of EDL’s employees engaged in collection. Approximately $1.6 million could be saved in years 2, 3, 4, and 5 of the program. About half of that amount should be realized in the first year, when the new collection process has been adopted. 6.3. Connecting a new customer Revised procedures for connecting new customers and reallocation of existing services have been recommended to EDL. By imposing the burden on the customer to inform EDL of a planned move, certain direct costs (crew dispatches) and indirect costs (office workers handling deposits, etc.) can be saved. In summary, the savings for this process are estimated to be $45,000 for the first year and $99,000 for each of the following years (2–5). 6.4. Begin installation of new meters with RF capability This step in the evolution of EDL to automatic meter reading includes replacing existing electromechanical (EM) meters with new ones (EM or electronic) equipped with RF communication modules. The new meters are read via radio signals using the same handheld units described in Section 6.1. The anticipated benefits include: Increased meter readers’ productivity leading to an opportunity to re-assign some meter readers to other duties, such as reading the RF meters every 30 days, collections or identification of energy-diversion activities. Ability to bill monthly those customers with RF meters to achieve a better cash flow and reduce interest charges on working capital. As a result of implementing this procedure, the following savings will be realized in years 2–5 respectively: $241,000, $554,000, $984,000, and $1,499,000.
34
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
6.5. Implementing vehicle-based meter readers This AMR option differs from the previous RF radio-based approach only in the use of a vehicle-based radio transmitter/receiver unit instead of the handheld units. Typically, the human reader will download the necessary data from the regional computer into the portable units employed in the vehicle-based system, and progress along a designated route stopping at strategic locations to read the meters. This optional step, if taken, would yield the following two benefits: More customers could be put on monthly billing as soon as the vehicle(s) are deployed, thereby saving additional carrying charges for receivables. Re-deployment of meter readers will no longer be required due to the use of the mobile reading fleet would permit applying additional attention to customer service calls, collection and auditing. The value of this benefit is the reduced personnel time devoted to resolving customer complaints. No dollar credit has been allocated to this option as it depends on when the mobile fleet is deployed. 6.6. Implement procedures for problem customers This section relates to the on-going program that EDL has for improving the recovery of lost revenues caused by various illegal measures. The steps described in Sections 6.1–6.5 should enhance EDL’s ability to identify the causes of non-technical losses and implement measures to recover the revenues represented by those losses. It is assumed that EDL will achieve recovery of up to 95% of the suspected $225 million of annual lost revenues in the fifth year of the program. The remaining 5% may take longer to detect and recover. This is an aggressive recovery-program, but should be achievable with concerted efforts by EDL, complemented by the modern metering and billing program proposed in this paper. Over the 4 years during which the RF equipped meters are implemented, it is estimated that EDL will recognize additional revenues of $30 million, $80 million, $150 million, and $214 million in years 2–5, respectively, of the program. A summary of the estimated benefits associated with the six steps defined in the phased implementation schedule and the total benefits of the project are presented in Table 3. It is clear from this table that, unlike the linear implementation of RF meters, this non-linear progression of revenue recovery represents a slow start but rapidly escalates as the penalties for diversion become well known and deter such future activities.
7. Cost/benefit analysis This section summarizes the results of the cost/benefit analysis for the modernization project using the following three alternatives. The third alternative is
35
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37 Table 3 Benefits by category and year realized Category of savings
$ millions
Introduction of handheld computers Change collection system Connecting a new customer Install new meters with RF capability Implement mobile AMR Implement procedures for problem customers Total
Year 1
Year 2
Year 3
Year 4
Year 5
1.748 0.788 0.050
3.496 1.575 0.099 0.241 0.0 30.0 35.411
3.496 1.575 0.099 0.554 0.0 80.0 85.724
3.496 1.575 0.099 0.984 0.0 150.0 156.154
3.496 1.575 0.099 1.499 0.0 214.0 220.669
2.585
recommended, but costs for options 1 and 2 are also presented to show the full range of the potential cost of the modernized system. Alternative 1: All electromechanical meters with RF communication modules. Alternative 2: All electronic meters with RF communication modules. Alternative 3: A combination of electromechanical and electronic meters equipped with RF communication modules. The cost presented for this recommended alternative is based on 14% electronic meters and 86% electromechanical meters. The electronic meters would be installed in year 2 for large consumption and problematic customers. The annual total benefits and costs for the three alternatives are presented in Table 4. It is clear from this table that recovery of revenues for diverted energy is the main opportunity for savings. Table 5 shows how the costs and benefits accumulate during the 5-year program. For the recommended system (alternative 3), the benefits overtake the costs in about 2 years. Any delay or slowdown in the revenue-recovery program will extend the time to achieve total recovery of revenues due to non-technical losses. Based on the presented recovery plan, a pay-back period of 2.7 years is possible for the recommended alternative. The pay-back period is defined as the period where the benefits of adopting the modernized system, when measured in dollars, Table 4 Annual cost and benefits for the three alternatives Year
1 2 3 4 5
Annual costs/benefits ($ millions) Savings
Cost (Alt 1)
Cost (Alt 2)
Cost (Alt 3)
2.58 35.41 85.72 156.15 220.67
1.04 20.87 20.87 20.87 20.87
1.04 47.75 47.75 47.75 47.75
1.04 35.92 35.92 35.92 35.92
36
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
Table 5 Accumulated cost and benefits for the three alternatives Year
1 2 3 4 5
Accumulated costs/benefits ($ millions) Savings
Cost (Alt 1)
Cost (Alt 2)
Cost (Alt 3)
2.58 37.99 123.72 279.87 500.54
1.04 21.91 42.78 63.65 84.52
1.04 48.78 96.53 144.28 192.02
1.04 36.96 57.83 78.70 99.57
equal the system’s expected implementation costs. The key to achieving this payback is to aggressively pursue the non-technical losses.
8. Conclusion This paper presented the results of a study for modernizing the electricity metering, billing and collection systems for Electricite´ du Liban, the national power company of Lebanon. Several types of meter reading systems were investigated for their applicability to Lebanon, including mobile, fixed, and power line carrier systems. The system recommended to EDL is a mobile AMR system with meter reading and data collection centralized at EDL’s headquarters and interfaced with the current billing-system. This system was identified as the ideal system for EDL because it is the least expensive, simple to install, quickest to implement, easiest to upgrade, compatible with other AMR systems, and field proven. In addition to the AMR system, the current meter-reading, billing, and revenuecollection processes at EDL were reviewed and new ones were proposed to provide EDL with better control of its current situation and to properly position the utility for the transition into the new automated system. A phased-implementation plan was also developed, based on a 5-year implementation period. Costs for the recommended system were obtained from several equipment suppliers and estimates of the accrued benefits over the implementation period of five years were developed. A cost benefit/analysis was then conducted for the modernized system. It was determined that this system is economically and technically feasible with a pay-back period of 2.7 years.
References [1] Appleyard D. Power theft: an insidious menace: Power Economics, July 1999. [2] Abi Said C, Baroudi R. A case study on Lebanese electric-power sector proposed restructuring and privatization program: the potential benefits for the Lebanon. Beirut Lebanon: Dar Al-Fan Lil-Tajlid; 2001. [3] EDL Business Plan: 1996–2000, EDL report, December 1996.
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
37
[4] Abi Said C. Midterm review: power sector restructuring and transmission expansion project. EDL report, June 1999. [5] EDL Report, Feasibility study for the EDL billing and revenue–collection modernization project. December 1999. [6] Scott HA, Schlenger DL. AMR today: sorting facts from fiction. Public Utilities Fortnightly, May 2000. [7] Scott HA. AMR deployments in North America. Power Value Magazine, November/December, 1998. [8] Scott HA. The benefits of automatic meter reading. The US Water Report, Saringa Group, October 1999.
Cost/benefit analysis of an AMR system to reduce electricity theft and maximize revenues for E´lectricite´ du Liban Raymond F. Ghajar*, Joseph Khalife Lebanese American University, PO Box 36, Byblos, Lebanon Accepted 22 January 2003
Abstract One of the largest pitfalls for any distribution network is the level of energy losses suffered by the system. These losses fall into two categories: technical and non-technical. Technical losses depend largely on the physical properties of the network, while non-technical losses (sometimes a more significant form of losses) are the result of theft or fraud caused by meter tampering, false reading, illegal connections or unpaid bills. In Lebanon, the levels of total losses are around 50% resulting in an annual deficit of more than 225 million US dollars. Despite the frequent breakdowns of the system and evidently unsustainable financial-losses, political consideration makes the sustained pursuit of electricity thieves low on the list of priorities. To overcome these hurdles, the national electricity company in Lebanon, E´lectricite´ du Liban (EDL), studied the possibility of using automatic meter-reading (AMR) technology to modernize electricity metering, billing and collection, minimize fraud and maximize revenues. The results of this study and a cost/benefit analysis of the proposed system are summarized in this paper. # 2003 Elsevier Ltd. All rights reserved. Keywords: Automatic meter-reading; Energy economics and management; Case study
1. Introduction The distribution of electricity has long been seen as a safe business venture. With secured earnings and a captive market, there appears to be little that could cast a * Corresponding author. Tel.: +961-9-547254; fax: +961-9-547256. E-mail addresses: [email protected] (R. F. Ghajar), [email protected] (J. Khalife). 0306-2619/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0306-2619(03)00044-8
26
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
shadow over such an investment. However, despite the apparently rosy glow of distribution network finances, the spectre of electricity theft lurks in the shadows of many developing countries, seriously impinging on their fundamental economics [1]. One of the largest pitfalls for any distribution network is the level of energy losses suffered by the system. Distribution losses depend largely on the physical properties of the network, such as conductivity and resistance of the cables, the number of voltage transformations made and the size of the network. The second, and sometimes more significant, form of losses comes under the heading of non-technical losses (i.e. theft and fraud). Several types of non-technical losses exist:
Fraud (meter tampering, false reading, etc.) Theft (illegal network connections) Collection (failure to collect monies owed) Billing (companies do not know how much electricity has been consumed).
All of these factors are exacerbated by the fact that detection of theft from a distribution network is extremely hard to identify. Essentially the only way to determine the level of losses, both technical and non-technical, is to subtract the level of payment for electricity delivered across the network from the value of power that was generated, the net figure thus produced gives an indication of the total losses. Attempts to determine empirical values for theft are, more often than not, simply estimates based on levels of detection. As Lebanon was being rebuilt after 20 years of civil war, the national powercompany, EDL, started to progressively modernize its infrastructures to meet the growing demand of residential, commercial and industrial customers [2–4]. With new generation and transmission facilities already built, EDL is currently attempting to improve its electricity metering; billing and revenue collection in order to increase revenues and reduce fraud, which accounts for almost 50% of missed revenues today [5]. The objective of this paper is to present a cost/benefit analysis of the proposed system for modernizing the electricity metering, billing and collection processes for EDL customers. Currently, these customers are being metered using a variety of old and new electromechanical meters. The metering process is manual, prone to errors and does not provide EDL with reliable methods to readily check bills, identify problematic customers, and reduce fraud. Several types of meter-reading systems were investigated for their applicability to Lebanon, including mobile, fixed and power-line carrier systems. The proposed system will ensure that EDL’s metering activities would become as advanced as those of modern utilities, reduce fraud and have a reasonable pay-back period. Most importantly, with the proposed system, EDL will regain the confidence of its customers, and would become a highly reputable and revenue-generating organization. The results presented in this paper will show that the proposed modernization project is technically and economically feasible.
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
27
2. Electric-metering solutions Electric and water utilities around the world are progressively moving towards automatic meter reading (AMR) technology to provide value-added services to their customers [6–8]. Most AMR deployments involve mobile AMR technology, where the meter reader uses handheld computers to communicate with the meter from a short distance. Drive-by solutions are also popular, where the meter-reading computer is installed in a van and the meter reader does not have to walk. The benefits of a mobile AMR for electricity customers are: 1. Faster reading of the meter, allowing for the reduction of the number of meter readers and reduction of operating costs. 2. Better accuracy of reading, allowing for the reduction of bill complaints. 3. Fewer contact with adverse elements (dogs, inclement weather, etc.) for driveby solutions. The additional benefits of a fixed AMR for the same customers are: 1. More frequent reading, for customers whose daily consumption is important. 2. More complex billing rates, allowing for incentives and marketing actions to retain and gain more customers where competition exists at the retail level. 3. Total elimination of the meter reader, in particular for difficult areas (high crime, very remote, hard to access, etc.). 4. Possible control of electric demand and the ability to perform on-request meter reading. The system recommended to EDL is a mobile automatic meter reading (AMR) system with radio frequency (RF) communication because it is the least expensive, simple to install, quickest to implement, easiest to upgrade, compatible with other AMR systems and field proven [5]. The equipment that is needed for the automatic metering system includes meters, communication systems, network-management software as well as data collection and storage software.
3. Technological alternatives These rely on a mobile automated meter reading system, which centralizes the meter reading and data collection at EDL’s headquarters and interfaces with the current billing system. This solution will give a proper incremental modernization that is compatible with EDL’s current financial and human resources. The proposed AMR system will include the following components: 1. Handheld computers that will be used in high-density areas for entering meter data and acknowledging payment of bills. Each unit communicates with the central server using a data network via a personal computer (PC) and a
28
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
2.
3.
4.
5.
6.
modem. The unit can communicate with the meter using short-distance radio and it also includes a keyboard for manual data-entry. Vehicle-based units will be used in low-density areas for reading meter data and acknowledging payment of bills. Each unit communicates with the central server using data network via a modem. The unit can communicate with the meter using a short-distance radio; it can also be used manually. Metering and route-management software, which will guide and control the meter reader data entry. The software will be installed in the handheld units and vehicle-based units. Collection-management software, which will control the collection route and the data entry for acknowledgment of bill payments. The software will be installed in the handheld units and vehicle-based units. Central management software installed at the central server, which stores the meter and collection data and interfaces with the current billing system. This interface will allow for daily transactions with the billing system, including updates of customer accounts, billing process, etc. New meter equipment, depending on the types of customers: For low-consumption customers (i.e. residential customers), the meters will be left unchanged. Data will be manually entered in the handheld units or vehicle-based units. For high-consumption customers (i.e. mostly commercial customers at the low-voltage level), the meters will be replaced with new ones (electromechanical or electronic) equipped with short-distance radio communication modules. The handheld units and vehicle-based units will communicate with the meters remotely, thereby eliminating all physical contact with the meter and reducing reading errors to a minimum. For problematic customers (i.e. residents who are tampering with the meters, pirating electricity, etc.), an electronic meter will be installed and EDL should consider enforcing the existing regulations to better deal with such customers.
4. Projected costs A summary of the capital and operating cost requirements for 943,000 customers (estimated number of EDL customers), for Alternative 1, electromechanical (EM) meters equipped with RF communication modules, and Alternative 2, electronic meters equipped with RF communication modules are summarized in Table 1. These cost estimates are based on budgetary-cost information gathered from suppliers of metering and billing systems. The capital requirement projections for both alternatives (electromechanical and electronic) are presented in Table 2. These projections are based on purchasing all 135 handheld computers, the regional host computers, and associated data collection software during the first year of implementation. Additionally, the
29
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37 Table 1 Budgetary cost estimates for the proposed AMR solutions Capital costs
Cost ($/item) Electronic
A. Component costs Meter w/RF Handheld computers w/RF Regional host computer Software: reading/scheduling/transfer
200
Quantity
Total
EM
Electronic
EM
86 943,000 4900 135 2000 12 200,000 1
$188,600,000 $661,500 $24,000 $200,000 $189,485,500
$81,098,000 $661,500 $24,000 $200,000 $81,983,500
Total
Total
B. Training costs Cost ($/h) Meter readers Supervisors/schedulers
4 4
Quantity 2160 480
$8,600 $1,900 $10,500
$8,600 $1,900 $10,500
C. System setup costs Cost ($/h) Data entry Network engineer Data analyst Programmer Meter installation
5 9 9 9 4
Quantity
Total
Total
1044 348 348 348 471,500
$5,200 $3,100 $3,100 $3,100 $1,886,000 $1,900,500 Total capital costs $191,396,500
$5,200 $3,100 $3,100 $3,100 $1,886,000 $1,900,500 $83,894,500
Cost ($/h)
Total
D. Optional components Cost ($/item) Billing system Remote disconnect Drive-by metering Advanced drive-by metering Spot billing Annual operating costs A. Operating costs Network Engineer Data Analyst Programmer
1,200,000 25 15,000 130,000 3800
9 9 9
Quantity 2088 2088 2088
$19,000 $19,000 $19,000 $57,000
B. Maintenance costs % Capital Cost Meters Handheld computers Host computers Billing system
0 10 10 0
Capital Cost 81,098,000 661,500 24,000 1,200,000
Total $0 $66,150 $2,400 $0 $68,550
30
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
Table 2 Capital requirements for the proposed AMR solutions Meters
Year Year Year Year Year
1 2 3 4 5
Handheld Host System computers computer set-up systems
EM
Electronic
$0 $20,274,500 $20,274,500 $20,274,500 $20,274,500
$0 $661,500 $47,150,000 $0 $47,150,000 $0 $47,150,000 $0 $47,150,000 $0
$234,500 $0 $0 $0 $0
$14,500 $471,500 $471,500 $471,500 $471,500
Operating & Annual capital requirement maintenance
$125,550 $125,550 $125,550 $125,550 $125,550
Total annual capital requirements
EM
Electronic
$1,036,050 $20,871,550 $20,871,550 $20,871,550 $20,871,550
$1,036,050 $47,747,050 $47,747,050 $47,747,050 $47,747,050
$84,522,550
$192,024,250
annual operating-cost for maintenance of the computers, and the fully-loaded cost for three information technology professionals to support the data collection system are included in the first-year capital requirements. In years two through to five, the capital and installation costs of 236,000 meters per year (i.e. 943,000/4) are included in the capital requirements projection, plus the annual operating cost. In addition to alternatives 1 and 2, a hybrid solution (the recommended alternative) that includes the following components is also proposed: New meters equipped with RF communication modules (943,000; 86% electro-mechanical and 14% electronic meters). Handheld computers capable of reading up to 2000 meters per day (135 units). Regional host computers (12 units). Metering and route management software. Portable meter reading computer capable of reading up to 4000 meters per day with spread spectrum (optional). Vehicle meter reading computer capable of reading up to 25,000 meters per day with spread spectrum (optional). The cost figures for the hybrid solution are based on the pro-rated costs of the previous alternatives. The recommended alternative gives the electric utility the freedom to choose the appropriate metering technology for each customer depending on his status (e.g. problematic, large consumption, regular, etc.). In addition to the proposed technology alternatives, a number of new processes were proposed to provide EDL with better control of its current situation and properly position the utility for the transition into the new automated system. Since these processes have an impact on the benefit analysis described later in this paper, they are briefly described below:
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
31
Separation of meter-reading and bill-collection processes. Use of handheld computers for meter reading and validation of customer information. Enhanced classification and identification of customers. Placing responsibility on the customer to pay the bill and not on EDL to collect. Inclusion of separate receipt with each bill. Usage of electronic meters to verify suspected problematic-customers. Imposing strict penalties on problematic customers. Disconnection of non-paying customers. Set-up of regional collection centres or coordination with banks as collection centres. Providing incentives for customers to pay their bills with automatic bank drafts.
5. Schedule for phased implementation A project plan for the phased implementation of the modernized system is presented in this section. The recommended modernization effort consists of six activities: Step 1. Purchase and introduce hand-held computers in year 1. Step 2. Modify the billing and collection processes, revise the customers’ database, and conduct an advertising campaign to inform customers with procedures for bill payment and penalties in year 1. Step 3. Install new meters with RF communication modules in years 2 through 5, targeting high consumption and problematic customers first. Step 4. Introduce one vehicle-based meter reader in year 2 to free some of the meter readers to perform other activities (optional). Step 5. Begin executing the recommended procedures for dealing with problematic customers in Year 2. Step 6. Upgrade the billing system in year 4 or 5 to meet EDL’s growing requirements (optional). The schedule for this complete modernization effort as illustrated in Fig. 1 can be completed in 5 years.
6. Projected benefits The benefits to EDL of modernizing its electricity metering, billing and collection processes are described in this section. The benefits are identified in terms of the steps identified in the phased-implementation schedule. While it is possible to
32
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
Fig. 1. Five-year plan for modernizing the metering, billing and collection systems by EDL.
accurately estimate the costs for implementing an AMR system, placing a dollar value on the potential benefits is far more difficult. The following discussion of potential benefits attempts to identify the nature of the benefits and, where possible, provides some measure of those benefits in terms of dollars, either saved or collected. 6.1. Introduction of hand-held computers This first step involves the issuing of handheld (computer) units to EDL meter readers to replace the manual recording of meter readings. It also requires that a system of regional computers be deployed for communicating between the handheld computers and the existing billing software that resides in EDL’s central mainframecomputer. Existing meters are not replaced at this point and the following three basic benefits are noted: Fewer customer complaints and more timely payments are expected. Using handheld computers that communicate directly to the billing program should reduce the number of data entry errors that find their way into the billing system. This should improve customer confidence of the resulting billings and lead to more timely collections. Fewer customer complaints of billing anomalies should result because of the more accurate readings obtained with the hand-held computers. Holding meter readers more accountable will reduce fraudulent reading practices. The high/low value checking feature of the hand-held units and the promise of an appropriate monetary penalty should thwart these practices. The projected benefits of employing handheld computers combined with penalizing the problematic customers should amount to approximately $3.5 million in years
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
33
2, 3, 4, and 5 of the program [5]. Considering a start-up delay, it is expected that about half of that amount could be realized in the first year of the program. 6.2. Change collection system Under the current system, an EDL employee delivers the bill personally to the customer and collects payment. This system places the responsibility of receiving payment on EDL. It often requires multiple trips to the same customer and results in a slow, time-consuming process. This responsibility needs to be reversed, placing the responsibility on customers to pay their bills in a timely fashion. A number of changes were recommended in bill processing including changes in bill delivery and revenue collection. These changes should yield the following benefits. Quicker collections of the metered portion of the consumed energy. Elimination of the need for many, if not all, contract collectors and freeing up some of EDL’s employees engaged in collection. Approximately $1.6 million could be saved in years 2, 3, 4, and 5 of the program. About half of that amount should be realized in the first year, when the new collection process has been adopted. 6.3. Connecting a new customer Revised procedures for connecting new customers and reallocation of existing services have been recommended to EDL. By imposing the burden on the customer to inform EDL of a planned move, certain direct costs (crew dispatches) and indirect costs (office workers handling deposits, etc.) can be saved. In summary, the savings for this process are estimated to be $45,000 for the first year and $99,000 for each of the following years (2–5). 6.4. Begin installation of new meters with RF capability This step in the evolution of EDL to automatic meter reading includes replacing existing electromechanical (EM) meters with new ones (EM or electronic) equipped with RF communication modules. The new meters are read via radio signals using the same handheld units described in Section 6.1. The anticipated benefits include: Increased meter readers’ productivity leading to an opportunity to re-assign some meter readers to other duties, such as reading the RF meters every 30 days, collections or identification of energy-diversion activities. Ability to bill monthly those customers with RF meters to achieve a better cash flow and reduce interest charges on working capital. As a result of implementing this procedure, the following savings will be realized in years 2–5 respectively: $241,000, $554,000, $984,000, and $1,499,000.
34
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
6.5. Implementing vehicle-based meter readers This AMR option differs from the previous RF radio-based approach only in the use of a vehicle-based radio transmitter/receiver unit instead of the handheld units. Typically, the human reader will download the necessary data from the regional computer into the portable units employed in the vehicle-based system, and progress along a designated route stopping at strategic locations to read the meters. This optional step, if taken, would yield the following two benefits: More customers could be put on monthly billing as soon as the vehicle(s) are deployed, thereby saving additional carrying charges for receivables. Re-deployment of meter readers will no longer be required due to the use of the mobile reading fleet would permit applying additional attention to customer service calls, collection and auditing. The value of this benefit is the reduced personnel time devoted to resolving customer complaints. No dollar credit has been allocated to this option as it depends on when the mobile fleet is deployed. 6.6. Implement procedures for problem customers This section relates to the on-going program that EDL has for improving the recovery of lost revenues caused by various illegal measures. The steps described in Sections 6.1–6.5 should enhance EDL’s ability to identify the causes of non-technical losses and implement measures to recover the revenues represented by those losses. It is assumed that EDL will achieve recovery of up to 95% of the suspected $225 million of annual lost revenues in the fifth year of the program. The remaining 5% may take longer to detect and recover. This is an aggressive recovery-program, but should be achievable with concerted efforts by EDL, complemented by the modern metering and billing program proposed in this paper. Over the 4 years during which the RF equipped meters are implemented, it is estimated that EDL will recognize additional revenues of $30 million, $80 million, $150 million, and $214 million in years 2–5, respectively, of the program. A summary of the estimated benefits associated with the six steps defined in the phased implementation schedule and the total benefits of the project are presented in Table 3. It is clear from this table that, unlike the linear implementation of RF meters, this non-linear progression of revenue recovery represents a slow start but rapidly escalates as the penalties for diversion become well known and deter such future activities.
7. Cost/benefit analysis This section summarizes the results of the cost/benefit analysis for the modernization project using the following three alternatives. The third alternative is
35
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37 Table 3 Benefits by category and year realized Category of savings
$ millions
Introduction of handheld computers Change collection system Connecting a new customer Install new meters with RF capability Implement mobile AMR Implement procedures for problem customers Total
Year 1
Year 2
Year 3
Year 4
Year 5
1.748 0.788 0.050
3.496 1.575 0.099 0.241 0.0 30.0 35.411
3.496 1.575 0.099 0.554 0.0 80.0 85.724
3.496 1.575 0.099 0.984 0.0 150.0 156.154
3.496 1.575 0.099 1.499 0.0 214.0 220.669
2.585
recommended, but costs for options 1 and 2 are also presented to show the full range of the potential cost of the modernized system. Alternative 1: All electromechanical meters with RF communication modules. Alternative 2: All electronic meters with RF communication modules. Alternative 3: A combination of electromechanical and electronic meters equipped with RF communication modules. The cost presented for this recommended alternative is based on 14% electronic meters and 86% electromechanical meters. The electronic meters would be installed in year 2 for large consumption and problematic customers. The annual total benefits and costs for the three alternatives are presented in Table 4. It is clear from this table that recovery of revenues for diverted energy is the main opportunity for savings. Table 5 shows how the costs and benefits accumulate during the 5-year program. For the recommended system (alternative 3), the benefits overtake the costs in about 2 years. Any delay or slowdown in the revenue-recovery program will extend the time to achieve total recovery of revenues due to non-technical losses. Based on the presented recovery plan, a pay-back period of 2.7 years is possible for the recommended alternative. The pay-back period is defined as the period where the benefits of adopting the modernized system, when measured in dollars, Table 4 Annual cost and benefits for the three alternatives Year
1 2 3 4 5
Annual costs/benefits ($ millions) Savings
Cost (Alt 1)
Cost (Alt 2)
Cost (Alt 3)
2.58 35.41 85.72 156.15 220.67
1.04 20.87 20.87 20.87 20.87
1.04 47.75 47.75 47.75 47.75
1.04 35.92 35.92 35.92 35.92
36
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
Table 5 Accumulated cost and benefits for the three alternatives Year
1 2 3 4 5
Accumulated costs/benefits ($ millions) Savings
Cost (Alt 1)
Cost (Alt 2)
Cost (Alt 3)
2.58 37.99 123.72 279.87 500.54
1.04 21.91 42.78 63.65 84.52
1.04 48.78 96.53 144.28 192.02
1.04 36.96 57.83 78.70 99.57
equal the system’s expected implementation costs. The key to achieving this payback is to aggressively pursue the non-technical losses.
8. Conclusion This paper presented the results of a study for modernizing the electricity metering, billing and collection systems for Electricite´ du Liban, the national power company of Lebanon. Several types of meter reading systems were investigated for their applicability to Lebanon, including mobile, fixed, and power line carrier systems. The system recommended to EDL is a mobile AMR system with meter reading and data collection centralized at EDL’s headquarters and interfaced with the current billing-system. This system was identified as the ideal system for EDL because it is the least expensive, simple to install, quickest to implement, easiest to upgrade, compatible with other AMR systems, and field proven. In addition to the AMR system, the current meter-reading, billing, and revenuecollection processes at EDL were reviewed and new ones were proposed to provide EDL with better control of its current situation and to properly position the utility for the transition into the new automated system. A phased-implementation plan was also developed, based on a 5-year implementation period. Costs for the recommended system were obtained from several equipment suppliers and estimates of the accrued benefits over the implementation period of five years were developed. A cost benefit/analysis was then conducted for the modernized system. It was determined that this system is economically and technically feasible with a pay-back period of 2.7 years.
References [1] Appleyard D. Power theft: an insidious menace: Power Economics, July 1999. [2] Abi Said C, Baroudi R. A case study on Lebanese electric-power sector proposed restructuring and privatization program: the potential benefits for the Lebanon. Beirut Lebanon: Dar Al-Fan Lil-Tajlid; 2001. [3] EDL Business Plan: 1996–2000, EDL report, December 1996.
R.F. Ghajar, J. Khalife / Applied Energy 76 (2003) 25–37
37
[4] Abi Said C. Midterm review: power sector restructuring and transmission expansion project. EDL report, June 1999. [5] EDL Report, Feasibility study for the EDL billing and revenue–collection modernization project. December 1999. [6] Scott HA, Schlenger DL. AMR today: sorting facts from fiction. Public Utilities Fortnightly, May 2000. [7] Scott HA. AMR deployments in North America. Power Value Magazine, November/December, 1998. [8] Scott HA. The benefits of automatic meter reading. The US Water Report, Saringa Group, October 1999.