On solving public facility planning problems using general mixed-integer programming methods

In many countries civil engineers play a fundamental role in the processes through which the location and capacities of public facilities like medical centers, schools, sanitary landfills, libraries, swimming-pool, etc., are decided. This paper describes a study aimed at evaluating the aptitude of general methods of mixed-integer programming (MIP) for solving public facility planning problems using PCs. The application of these methods is quite easy today, even for people with limited optimization and computing skills, thanks to the good, friendly MIP packages now available. The evaluation was performed on a collection of basic problem types involving both minimum-cost and maximum-accessibility objectives. The main, general conclusion of the study was that, at present, problems with up to 100 centers and 100 sites, like most real-world problems are, can be solved rather efficiently on many occasions.