skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Forecasting for energy and chemical decision analysis

Abstract

This paper focuses on uncertainty and bias in forecasts used for major energy and chemical investment decisions. Probability methods for characterizing uncertainty in the forecast are reviewed. Sources of forecasting bias are classified based on the results of relevant psychology research. Examples are drawn from the energy and chemical industry to illustrate the value of explicit characterization of uncertainty and reduction of bias in forecasts.

Authors:
Publication Date:
Research Org.:
Decision Focus Inc., Los Altos, CA
OSTI Identifier:
5765574
Report Number(s):
CONF-840828-
Journal ID: CODEN: ACENC
Resource Type:
Conference
Resource Relation:
Journal Name: AIChE Natl. Meet.; (United States); Journal Volume: 49D; Conference: International synfuels policies and technologies symposium, Philadelphia, PA, USA, 19 Aug 1984
Country of Publication:
United States
Language:
English
Subject:
02 PETROLEUM; 03 NATURAL GAS; CHEMICAL INDUSTRY; DECISION MAKING; INVESTMENT; MARKET; FORECASTING; NATURAL GAS; NATURAL GAS INDUSTRY; PETROCHEMICALS; PETROLEUM; PETROLEUM INDUSTRY; ECONOMIC ANALYSIS; PROBABILISTIC ESTIMATION; SUPPLY AND DEMAND; ECONOMICS; ENERGY SOURCES; FLUIDS; FOSSIL FUELS; FUEL GAS; FUELS; GAS FUELS; GASES; INDUSTRY; PETROLEUM PRODUCTS 020700* -- Petroleum-- Economics, Industrial, & Business Aspects; 030600 -- Natural Gas-- Economic, Industrial, & Business Aspects

Citation Formats

Cazalet, E.G. Forecasting for energy and chemical decision analysis. United States: N. p., 1984. Web.
Cazalet, E.G. Forecasting for energy and chemical decision analysis. United States.
Cazalet, E.G. 1984. "Forecasting for energy and chemical decision analysis". United States. doi:.
@article{osti_5765574,
title = {Forecasting for energy and chemical decision analysis},
author = {Cazalet, E.G.},
abstractNote = {This paper focuses on uncertainty and bias in forecasts used for major energy and chemical investment decisions. Probability methods for characterizing uncertainty in the forecast are reviewed. Sources of forecasting bias are classified based on the results of relevant psychology research. Examples are drawn from the energy and chemical industry to illustrate the value of explicit characterization of uncertainty and reduction of bias in forecasts.},
doi = {},
journal = {AIChE Natl. Meet.; (United States)},
number = ,
volume = 49D,
place = {United States},
year = 1984,
month = 8
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • The increase in energy prices in 1973 and the energy embargo stimulated the application of public and private resources to analyze the determinants of energy use in the U.S. This paper reviews the pre-1973 modeling environment and summarizes the major modeling and analytical developments since that period. The major change in energy demand analysis in the ten years since 1973 has been the development of structural or engineering/processing models, that have replaced the reduced form models used in the post and early pre-1973 time period. Energy demand models have had three major policy uses in Federal agencies: forecasting, program analysis,more » and impact analysis. This paper outlines the development of energy demand models and their interaction with policy analysis and reviews what this experience has taught the energy modeling community.« less
  • Oil- and gas-supply modeling may not only generate forecasts on a stand-alone basis, but may provide input data and assumptions to large-scale, long-range integrated energy economy models. In such a framework, parameters that may not have seemed especially crucial in the stand-alone formulation may be found to have an unexpectedly great influence on the results. Additionally, it may become necessary to incorporate considerations that are usually omitted from conventional oil- and gas-supply models; this paper discusses two such examples: the decline rate, or more generally the production profile, and the treatment of resource exhaustion. This investigation arose in connection withmore » a project for the Energy Information Administration (EIA). One part of the project was to incorporate the assumptions and input data of EIA's Midterm Oil and Gas Modeling System (MOGSMS) into EIA's Long-Term Energy-Analysis Program (LEAP). Two important areas for oil- and gas-supply model development and data collection are representation of the decline rate or production schedule from reserves, and the treatment of resource depletion including the subeconomic resource base and enhanced recovery. Both aspects receive relatively little attention in oil- and gas-supply models, but play key roles when the assumptions of these models are integrated into a long-term model such as LEAP.« less
  • Knowledge engineering techniques were used to study the decision process for choosing which of a set of recommended energy conservation options would be implemented. Building management decision-makers from both the private and public sectors were interviewed to gain an understanding of the decision-making process. Decision objectives were identified and the process computerized. Results of the study are twofold. The first is a formalization of the decision-making process. The formalization enables both efficient treatment of large numbers of objectives and demonstration of optimality in meeting objectives. Second, the knowledge-based system produced is programmed in a conventional programming environment rather than amore » rule-based expert system shell, demonstrating the range of applicability of knowledge engineering techniques.« less
  • The array of alternative energy sources which are vying for the federal government's R and D dollar is formidable when compared to the politically acceptable amount which can be used to fund the research. To guide how these funds should be dispersed, a rational, defensible procedure is needed which can repeatedly be applied as new technologies and new information become available. The procedure advanced in this paper is a decision analysis technique known as multi attribute decision analysis (MADA) and its use is illustrated in an evaluation and ranking of solar thermal electric power generating systems. Since the ultimate purchasemore » decision is made in the market place, the preferences of potential users have been sampled and brought to bear on the ranking. The focus of this description is on the formulation of the problem structure and the decision model, the treatment of uncertainty, and how the results relate to the questions asked by and of the Department of Energy, which funded the study. A final note proposes how decision analysis can be used to address the broader questions of choice among competing technologies with cautions concerning misuse of the procedure.« less