|
|
|
|
Building Models for the “Smart Grid” Business Case
By Jagoron Mukherjee, Senior Consultant, KEMA
|
|
|
The Federal Energy Policy Act 2005, Section 1252 (EPAct) states that energy policy supports all utilities to “provide customers with time-based rates and the ability to receive and respond to electricity price signals.” The EPAct, however, does not specifically mention or require that utilities do anything directly, but instead it requires the regulators and the Boards of Directors of unregulated utilities to “consider and determine” what these utilities must do to comply with the objectives of the EPAct. This regulatory driver, in tandem with recent developments in communication and Information Technology (IT) and increased cost of “clean” conventional energy sources, have created an opportune environment to seriously consider technologies such as smart meters, Advanced Metering Infrastructure (AMI), and “smart grid” as practical solutions to address power delivery needs of the future. Consequently, many leading utilities in the United States have already commenced with some form of these programs and many more are being designed and specified. In this article, based on KEMA’s recent experiences with several smart grid initiatives, we present an overview of the purpose for building a business case for the smart grid, the role of financial models in the business case, and why a generic “one-model-fits-all-needs” approach to modeling smart grid may have limitations.
|
|
|
Business Case for Smart Grid
Whatever the primary objective and key drivers for a utility may be, a key step for most utilities is to closely examine the economic value of implementing a smart grid solution before committing to an actual deployment. These deployments are capital intensive and impact multiple departments within the organization, as well as directly impact the cost of energy and service levels for the end-use customer. As with most capital budgeting and economic value analyses, the conventional approach is to project the economic value of an investment and then assemble a “business case.”
The business case is a useful way for stakeholders to capture smart grid capabilities which typically consist of everything needed to support a) Advanced metering, b) AMI communications networks, c) Home or personal area networks, d) Distribution automation sensors and nodes, and e) Data systems and interfaces to legacy applications.
The business case for smart grid has emerged as powerful tool to project operational efficiencies obtained through the principal benefits of smart grid and then compare these, often incrementally, with the costs of multiple technology options and functionality to reveal which combination of capabilities have the greatest potential value. Typical operational benefits include (but are not limited to) reduced meter reading costs, reduced costs associated with field visits and customer calls, improved billing accuracy and improved cash flow, improved outage information and response, and more efficient asset management and distribution engineering design. Building a robust business case with sufficient research and validations, helps utility leadership move from an intuitive to a more analytical decision point, and fosters increased confidence in economic projections to support regulatory cost-recovery proceedings. Under the aegis of the Sarbanes-Oxley Act, as it applies to investor owned utilities, building a business case for large capital investments, such as the smart grid, with validated financial data and defensible assumptions also contributes to the culture of transparency and discipline required to record, process, summarize, and effectively report financial projections.
|
|
|
Role of Models in the Smart Grid Business Case
In developing the business case, the most common approach is to develop a financial model that is generally spreadsheet-driven. To quote the Nobel Prize winner Robert Solow, “model building … is one of the very important ways of knowing the modern world: forming a simplified picture of the way the world works, reckoning what sort of evidence is relevant to doing it right, and judging the consequences of alternative patterns of policy.”
To address this most common practice of building models for smart grid business cases, there is increased discussion in some parts of the industry to use a generic valuation model to support regulatory reviews. Proponents of this approach argue that there are overall merits in coming to common ground on different artifacts of valuation modeling, specifically in key cost and performance assumptions and reporting methodology.
It is commonly believed that such a “one-model-fits-all-needs” approach will greatly assist utilities and public utility commissions (PUCs) in providing a reference or a baseline to conduct a preliminary value assessment for the installation of smart grid and its supporting IT network and infrastructure. In recent months, we have observed that a number of state PUCs are requiring utilities in their jurisdictions to use such generic valuation models to develop and report their business cases as part of cost-recovery and approval requests. These examples have focused principally on the AMI component of the smart grid architecture, so these observations are similarly focused.
Those who are in favor of using a generic approach typically argue that such an approach provides good transparency to the valuation process. For instance, a common framework would be expected to uniformly drive all utilities to estimate the typical measures of a standard business case, such as net present value (NPV) of the investment, depreciation calculations, and the internal rate of return (IRR). However, we have found that such an approach is not practical. We would suggest that this approach to AMI or Smart Grid valuation is more of an exception than a rule. Consequently, a generic valuation model should be treated as a general framework which, in most cases, would require extensive refinements and customization to determine the value of an AMI deployment for a specific utility. In the section below, we provide few observed scenarios of smart grid valuation analyses to substantiate why using a singular approach can be limiting.
Regional Variations: Utility service area differences alone, such as urban/rural topographical distinctions, population density, customer demographics, and climatic impacts to energy consumption make certain AMI communications technologies more viable than others. In many business cases, the benefit derived from avoided or deferred capital costs are based on the assumption that demand response programs enabled by AMI can displace the need for new generation facilities, or at least reduce the procurement of incremental capacity in the market. Depending on regional conditions of service reliability criteria (SAIDI/SAIFI) , cost of building new generation, information requirements for customers, tax laws, and other important factors, the valuation modeling assumptions can vary significantly across utility service areas.
Multi-State Regulatory Requirements: While the EPAct has mentioned initiatives to modernize the grid, it is important to note that much of the actual operation of the grid as well as decisions on investments to the grid, with respect to generation, transmission, and especially on the local distribution grid, are under the jurisdiction of the States. States jurisdiction is conducted through state legislatures and/or public utility commissions – and not the Federal Government. These decision-making bodies have authority over generation, transmission, distribution, and demand-side energy efficiency and demand response management. Each state has its own objectives and, often, its regulatory requirements are quite distinct from one another. For example, some states have a mandated equipment testing and certification schedule for retiring existing assets, for installing new assets, as well as criteria for claims and force majeure events. All of these requirements will impact the costs and benefits for deploying smart grid technologies and need to be incorporated into the valuation exercise. In the case of utilities that have multi-state jurisdictions, the common modeling approach in one jurisdiction is not consistent with the regulatory requirements in its other areas, creating further complexity for the utility’s planners.
Diversity in Deployment: Utilities that pursue a smart grid deployment usually do so in multiple stages. Accordingly, different utilities, for that matter, different areas within a utility, will most likely be at varying stages in the overall deployment roadmap. A typical implementation may start with the automation of meter reading by installing smart meters and extends to communication and IT architecture, and finally towards a smart grid which consists of additional automation functionalities within the realm of the distribution network, and perhaps even the customer’s own personal network.
For instance, the distribution infrastructure includes substations and feeder circuits to carry power to neighborhoods and then distribution transformers steps down the voltage to our houses or commercial outfits. The AMI infrastructure integrates with the distribution system and provides automated meter reading and is augmented with an IT infrastructure. The most common benefits that are sought are through automated readings of kWh usage, i.e., energy consumption, and tamper detection, which are both directly used to support billing and revenue collection activities.
The smart grid is an extension over AMI. With sensors, instrumentation, and IT added to the substations and the lines themselves, massive amounts of data are collected and then processed so that actions can be taken in an automated manner. Using IT, it is possible to provide energy price information to customers thus providing an opportunity to optimize usage based on that data visibility, be it through manual or automated means. Subject to the prevailing interval price of power, the customer can then adjust usage. In case of power or gas outages, outage information is easily available and communicated to the user. In addition, using these sensors, utilities can automatically analyze all of that data for control purposes, for asset monitoring, for power-quality monitoring, and for increased outage intelligence. In summary, the smart grid allows utilities to take a proactive approach to outage detection and restoration rather than the other way around, i.e., relying on customers to alert utilities to power outages. The utility will know exactly where the outage is, what equipment is affected, and what the root cause is and automatically dispatch the repair crew. Additionally, the smart grid has the potential of isolating the fault with automatic switching and restoration of power service to as many customers as possible, by rerouting power flow around the problem.
As the deployment incrementally moves more towards a smart grid initiative from AMI, benefits from kW Interval Data, dispatchable rates, outage monitoring, read on-demand, selectable, billing dates, customer usage profiles, and dynamic load research are required to be included in the model to develop the complete business case. Evidently, a utility may have multiple areas under different stages of deployment or a particular utility may choose only to implement a subset of all the components of the smart grid for its own economic or strategic reasons. Therefore, there is considerable customization of the valuation model in estimating the timing and sequencing of these benefits and costs to reflect the true deployment diversity – all of which would be very hard to include in a simple generic valuation model.
In addition, key assumptions regarding improvements to utility operations will vary considerably across the range of services that be supported at various performance levels. For example, some utilities outsource all or portions of their meter reading, call center, or billing operations, thus the valuation of benefits related to these areas will require changes in formulation and timing, as well as inclusion of costs for such items as early contract termination.
Including System-wide Benefits with Utility Revenue Benefits: In recent months, many public utility commissions are requiring that utilities include system-wide benefits into their business case. This change is significant because utilities typically did not have the necessity, motivation, or incentive to evaluate and include benefits that would occur to customers and to the society at large and not directly impact the company’s balance sheet or income statement. Furthermore, the relative size of previous automated metering investments did not rely as heavily on enterprise-wide benefits, as they do today for AMI. Valuation models, therefore, often need to quantify societal benefits, such as avoided generation investment, reduction of greenhouse gases and overall carbon footprint, or intangible customer benefits such as increased satisfaction due to better service and billing, and wider service offerings and choices.
Some of these benefits, such as increased customer satisfaction, though hard to quantify, are benefits nevertheless and, depending on the regulatory environment, may need to be considered in the regulatory review process. The rationale to include these benefits is that despite the lack of realization of some of these societal or non-operational benefits, the market or society at large benefits from various aspects of implementing smart grid technologies and needs to be considered in these discussions.
Other Special Analyses: In most practical situations, the valuation modeling for an investment of this size and importance are often conducted in conjunction with other specialized analyses within the enterprise. Many utilities will need to consider the impacts to, or from, the AMI or Smart Grid analysis and may already have incremental processes to conduct these analyses. This strong coupling may requires the smart grid valuation modeling to be structured in a manner that supports other internal utility analyses, particularly as it relates to relevant financial metrics and other key output variables. Some of these additional analyses could include (but not be limited to) the following:
Capital Allocation Modeling: In practice, a large utility will be evaluating multiple investment options simultaneously; some of which will be competing for a limited amount of capital. Smart grid capital projects will likely occur over multiple financial and budget cycles. As a result, utility decision makers would be faced with choosing from an array of projects that may be funded over similar or much shorter budget cycles. In almost all cases, the choice of projects is based on an optimal decision to maximize the overall benefit subject to the constraints of limited finances, particularly where working capital recovery may be limited in the regulatory process. Under such situations, it becomes necessary for utilities to more carefully analyze AMI program funding as part of an optimal portfolio of investments, to maximize overall return on investment or meet similar metrics consistent with the utilities’ strategic, financial, and risk considerations.
Risk Analysis and Strategic Decisions Under Uncertainty: Smart grid deployments, like any other large scale projects (e.g., power plants), are faced with inherent uncertainties. In addition to usual project management uncertainties regarding project schedule, resource planning, and execution, uncertainties related to new product and technology performance can also have a significant impact on the business case outcome. Depending on the complexity of the deployment, conducting risk analysis and identifying sensitivities in costs and benefits to variation in key inputs may become important in the decision making. For example, energy demand elasticity usually has a variance. To meet the goal of resource adequacy, utilities may assume a certain demand response in deciding not to construct new facilities and, in the process, include avoided capital benefits in the financial model. However, in reality, a variation of demand response may occur, forcing construction of new facilities, which may vastly change the outcome of the business case.
To facilitate the risk analysis, usage of probabilistic risk assessments such as, Monte-Carlo simulation and other sophisticated valuation techniques (e.g., real-options) may need to be either incorporated into the model or performed post-modeling. While it may be argued that quantified cost-benefit analysis should not be the only consideration in deciding the merit of an investment case, it certainly has become the principal focus for evaluating smart grid investments and in deciding whether the investment is in the public interest. The costs and the potential benefits of these projects are inherently uncertain, and difficult to quantify, as is the case with any new technology and uncertainty in service level and customer acceptance. A robust and exhaustive model, with sufficient scenario analyses and probabilistic risk assessment, becomes a very effective tool to understand the quantitative implications of strategic decisions and thus an important aid in helping decision makers in utilities to make the best choices under all these uncertain considerations.
|
|
|
Conclusion
Building the business case is an integral part of the AMI/smart grid initiative in which many utilities are seeking to embark. The business case is vital for justifying the investment - internally - for large Investor Owned Utilities to ensure that such investments have economic merit. Externally, the business case provides the principal means to justify regulatory cost-recovery where these investments are to be included in revised rate structures. To facilitate this modeling effort and establish a common baseline, there is an increased interest in seeking generic frameworks and valuation models among regulators and other stakeholders. While this “one-model-fits-all” approach is useful in providing the broad requirements of the business case, it does not provide sufficient treatment of the specific requirements of the utility’s specific situation. Utility managers, regulators, and other potential users are advised to exercise a good understanding of their issues and conditions and then determine the modeling approach for their AMI/smart grid deployment with company-specific modeling tools and more specific assumptions and key inputs.
Contact the author at jagoron.mukherjee@kema.com.
|
|
|
Download the January 2008 Issue
Use the link below to download the PDF of the full issue of the January 2008 Automation Insight to view the complete print versions of the articles.
|
|
|
|
[download] Automation Insight Jan 2008 (.pdf 334 kb)
|
|
|
|
|
|
|
|
|
|
