Compering Linear Programming Tools and various B&B algorithms for the Resource Sharing and Scheduling Problem

The collaboration of robots or other renewable resources to efficiently perform complex operations is an essential part of modern industrial systems. The key features of the problem considered in this work include alternative modes for executing an operation and the specific set of collaborating resources, each of which may be needed for a portion of the operation duration, required while performing an operation under a certain mode. The objective of the problem is to minimize the makespan by choosing one mode for each operation (i.e. determining resources allocation and collaboration) and finding sequence and timing of operations (i.e. setting resources scheduling).

Several studies in the past were focus on developing an optimal branch and bound (B&B) and genetic algorithm heuristic to solve the RSSP. This work was focus on two issues: (1) examining the efficiency (runtime) of computing linear programming lower bounds via Lindo, CPLEX; (2) examining the influence on the efficiency of several customized B&B tree structures.