Solving the installation scheduling problem using mixed integer linear programming
R. Chen, Harlan P. Crowder, E. L. Johnson · IBM Systems Journal · 1978
We have defined the installation scheduling problem, formulated it as a mixed integer linear program, and have shown how to solve it using the Installation Optimization System. We have indicated that the solution techniques embodied in ios are applicable to wide classes of similar problems. There are extensions of the installation scheduling problem which we have not addressed but which might be interesting to pursue. For example, in many applications a substantial benefit is to be derived when the final action is performed. We have not investigated what effect, if any, such terminal benefits could have on optimal installation schedules. Another topic that could be significant in certain applications is the question of purchase versus lease of systems. We feel that investigators seeking solutions for these and other extensions of the installation scheduling problem should consider the mixed integer programming approach that we have presented