Evaluation of on-line schedules by distributed simulation

Abstract

A new algorithm for the distributed simulation and evaluation of on-line schedules is presented. Generally, on-line scheduling has often been restricted to scheduling activities on a single machine or workcell. The exploratory research reported in this paper expands on-line scheduling to encompass several machines or workcells. Branch and bound search techniques are used in the simulation to reduce the number of simulations simultaneously in execution. The algorithm is applied for the distributed simulation of on-line schedules for a manufacturing example. * This research was supported in part by the Texas Higher Education Coordinating Board under grant ATP-115. 1 Scheduling can be classified as either on-line or off-line scheduling. In this paper, scheduling is used to refer to on-line scheduling, unless otherwise explicitly stated. For more discussions on scheduling, the reader is referred to [FoKe85] and references therein. 2 A minimally constrained plan is a plan that will not require replanning in the future. Constraints known apriori are used to generate a minimally constrained plan off-line. A maximally constrained plan is a plan that is generated taking into consideration all available constraints. A maximally constrained plan is generated at run-time and can be directly executed. L. Camarinha-Matos et al. (eds.), Balanced Automation Systems