Activity-investment scheduling for maximizing cash availability subject to capital constraint

Abstract

This thesis considers an activity-investment scheduling problem for a project consisting of multiple activities where activities are precedence-related each other and each activity requires a certain amount of capital investment and results in either deterministic cash-in (positive) or cash-out (negative) flow at the completion of its investment. The objective is to find the optimal activity-investment schedule (planning) over a fixed planning horizon that maximizes the cash availability subject to activity precedence relations and capital limits. In the analysis, the activity-investment scheduling problem is transformed into a parallel machine scheduling problem which is formulated as an integer programming and proven to be NP-hard. To solve the problem, a depth-first branch-and-bound algorithm is derived. The numerical experiment shows that the algorithm solves the problem efficiently and effectively. This thesis may immediately be applied to various projects in the areas of construction, defense, finance and entertainment.