A two-step approach to planning new facilities towards equal accessibility

A recent advancement in location-allocation modeling proposes a new objective of minimizing inequality of accessibility. Existing work considers the planning problem as either selecting new sites or adjusting facility capacities, separately. This paper develops a two-step hybrid approach to the problem by optimizing both locations and capacities of facilities towards equal accessibility. A genetic algorithm is first employed to find the best locations to site new facilities, and then a quadratic programming method is used to determine the best capacity of each facility within a pre-defined range. The sequence is consistent with many decision-making practices. Results of a series of experiments demonstrate that location optimization reduces inequality in accessibility more significantly than capacity optimization. The two-step optimization method can be applied for sequential allocation decision-makings towards maximum equal accessibility.