Operation Simulation Associates
November 19, 2009
The staff and computer models of Operation Simulation Associates, Inc (OSA) have been involved in the electric utility Integrated Resource Planning Process (IRP) for more than thirty years. The purpose of this paper is to define the objectives and methodology for conducting an adequate IRP in the context of the new opportunities and risks facing the energy industry and its many stakeholders in today’s world. This paper will be followed by the development of a detailed IRP for a typical, but hypothetical, energy service area using the PowrSym3 model and an Internet based forum for stakeholder interaction with the process.
The stakeholders are everyone who will be affected by the decisions made in the planning process. The stakeholders vary greatly in their viewpoints and often have conflicting objectives. The size of the “stake” held by various stakeholders varies greatly. An important fact to remember is that more than half of the people whose quality of life may be affected by a large power plant project will be born after the decision has been made. Many of the stakeholders may not even live in the region under study as fuel supplies are often imported into the region and some of the emissions from power plants may affect the global environment. Let’s attempt a definition of a stakeholder:
A stakeholder in the Integrated Resource Planning Process is anyone whose quality of life may be affected by the projects envisioned in the IRP over the service life of these projects.
This definition certainly includes the owners of the projects, the fuel suppliers, and the residential and commercial consumers in the region. It may also include stakeholders outside the region who are affected by fuel supplies, waste disposal, or emissions of the projects.
Given the wide diversity of stakeholders with conflicting objectives, the only viable economic objective function is to minimize the total societal costs of meeting the energy needs of the region. The total societal costs include not only the directly incurred costs of the projects but also the external and future costs caused by the various projects over their entire life cycle, from the beginning of the study forward. It is important to note that the objective function does not include “sunk costs” that can not be avoided due to prior decisions. For example, every ton of coal ash added to the storage sites going forward is an avoidable cost that is a variable in the study, but the liability for the millions of tons of ash already in storage is a sunk cost.
In addition to incurred costs, there can be incurred external benefits. Again using the coal ash example, a project that reduces the amount of the old ash in storage reduces a liability and is thus a benefit.
The objective function is thus the minimization of the total societal costs, offset by any societal benefits created, for meeting the energy needs of the region.
An IRP typically begins with a base case extending twenty to forty years into the future with a “business as usual” growth plan. Then variations of the base case are created with various penetrations and combinations of alternate technologies. Note that the technologies to be considered include both energy supply and energy conservation, a watt saved is as good as a watt produced. The following list includes obvious options that must be considered, but is not meant to be all-inclusive.
All of the data for each alternative is subject to future risks and uncertainties. In addition there is large disagreement concerning some of the data, for example the societal damage caused by CO2 emissions. These factors combine to create a range of uncertainty for each data input, a range that increases as we look further into the future. Uncertainty is properly handled in an IRP by running many cases with different futures for many of the data inputs.
Note that the combination of many future options and treatment of data uncertainty will result in thousands of cases to be run through the computer models producing a myriad of results that are best analyzed graphically. This is not a quick or easy process and IRP studies that are limited to fewer cases are not likely to produce optimal results.
If sufficient cases are studied and agreement is reached on the data ranges, then an optimal strategy will be identified. But a more likely result is that different stakeholders will have different ranges of uncertainty for various data items. In this event, the result of the studies will be multiple strategies based on the views of the various stakeholders.
The purpose of the computer models is to provide an analytical “level playing field” upon which the various options and data range assumptions can be tested and compared. The computer models will find the lowest cost operation to serve the region’s energy needs for each scenario of options and data variations to be evaluated. The result will be a graphical representation that will allow identification of an optimal plan for each data outcome and, subject to the viewers’ assessment of data ranges, identify the risks associated with the selected plan.
An adequate IRP will require many computer cases and multiple interactions with the stakeholders. The completion of one group of cases will often lead to the suggestion of new cases. A typical IRP process requires about two years. By using the Internet for on-going collaboration with the stakeholders, it may be possible to compress the process to one year.
The CIRPS project differs from most utility sponsored IRP studies in objective functions, focus, stakeholder participation, and scope.