Abstract
This study deals with the development and application of a methodology which permits optimal planning of an urban streetlighting system. It presents for the first time—to the best of the authors' knowledge—a procedure for the quantitative estimation of the utility accruing from street-lighting. The multivariable utility function divided by the installation cost of lighting for each street yields an efficiency coefficient for a given lighting project. The rank order of these coefficients, listed in descending order, shows the marginal utility of each project realized.
An optimization model based on an integer programming algorithm was employed because it permits periodical (for example, annual) selection of a set of indivisible projects to be realized during a given period so as to derive maximum economic and social utility, subject to budgetary and technological constraints.
The first part of the methodological derivation yields the utility function of the lighting system for each street. The components of the function, some of which represent subjective assessments, were identified by the Delphi method, which permits derivation of subjective values for groups of interviewees whose assessments may reasonably be assumed to be crucial in determining the relative values of the objectives used in the utility function.
The optimization process was carried out using the IBM MPS X (an integer programming algorithm). The selection process was extended and deepened so as to render it sensitive to benefit from economies of scale and external economies. The conventional practice, whereby residents have no say in the planning of the lighting system in their city, was abandoned. Residents' value judgements concerning the relative importance of the objectives to be served by the lighting system were incorporated into the planning process.
Get full access to this article
View all access options for this article.