Optimizing and solving project scheduling problem for flexible networks with multiple routes in production environments

A. Birjandi | Seyed Meysam Mousavi | M. Hajirezaie | B. Vahdani

---

URL :   http://research.shahed.ac.ir/WSR/WebPages/Report/PaperView.aspx?PaperID=116882
Date :  2019/04/29
Publish in :    Journal of Quality Engineering and Production Optimization

Link :  http://jqepo.shahed.ac.ir/article_895.html
Keywords :Flexible production networks, RCPSP, Production projects, Production scheduling problem, Mathematical model, Meta-heuristic algorithm,Multiple routes

Abstract :

---

Resource-constrained project scheduling problem (RCPSP) with multiple routes, namely RCPSP-MR, is a type of RCPSP in production environments, in which operations and production activities are divided into two types of flexible and fixed parts. Each flexible activity has multiple routes for its execution. A route can be denoted as a semi-finished product with its production activities and precedence relations while requiring the same resources in an overall problem. In this paper, a new mixed‐integer nonlinear programming (MINLP) model is presented to minimize the makespan. The optimal value of the objective function is highly depended on an optimal route for each flexible production activity. Therefore, the proposed mathematical model identifies the optimal routes and consequently determines the optimal project network. Also, the model allocates renewable resources for each production activity. Finally, production sequencing of activities is optimized using the proposed model. To solve large-scale problems, a new hybrid heuristic based on genetic algorithm (GA) and binary particle swarm optimization (BPSO) is introduced to handle two main sub-problems of RCPSP-MR in production environments, namely, route selection and production scheduling. To assess the performance of the presented optimization model and proposed solving approach, 60 test problems in various sizes are reported in detail.