Optimal location and design of a competitive facility (Q1881559)

Given a finite set of existing facilities (e.g., customers) and of already operating competing facilities (e.g., stores), a profit-maximizing location of one new facility is sought. The expected profit of the new facility is assumed to be directly dependent on the captured market share, which in turn depends on the location and the quality of the new facility. Considering both the location of the new facility and its quality as decision variables, a general profit-maximizing competitive location model is formulated. It is shown that the problem is well-posed if customers are assumed to be novelty-oriented (i.e., if they prefer the new facility in case of ties with respect to attractiveness, a function of distance and quality), while other tie-breaking rules may lead to problems for which optimal solutions might not exist. For the first case, a reduction of the problem to an associated biobjective minquantile-maxcovering problem [see the authors, in Math. Program. 71, No. 1(A), 101-112 (1995; Zbl 0855.90113) for more details] is derived, the efficient solutions of which determine a candidate set for the optimal solution of the considered competitive location problem. For problems in the Euclidean plane and, e.g., gravity-type attraction functions, solution methods which are based on the determination of such a candidate set are polynomial in the number of existing facilities. This approach is illustrated at a small example problem, and extensions to other cases (i.e., distance and attraction functions) are also discussed.