Tackling the complexity of designing multiproduct multistage batch plants with parallel lines

Subtitle:the application of a cooperative optimization approach

Over the last decade, cooperative optimization approaches, that combine metaheuristics with exact methods, have been attracting a lot of interest to solve mixed integer linear programming (MILP) problems. In this paper, we present such a hybrid technique for the batch plant design problem with parallel production lines. The aim of this MILP problem is to minimize capital and setup costs through the optimization of 3 decisions: the number of lines to install on the plant, their design, and the assignment of products and production quantities to the installed lines. These different types of decisions lead to a natural separation of the global problem into subproblems, each of which can be solved with adequate solution methods. More specifically, we applied an iterative local search (ILS) metaheuristic for the discrete design decisions for every stage of every line, whereas the product-to-line assignments are optimized with an exact MILP solver. We present a case study of a lubricant plant to compare the performance of the proposed methodology to an exact method, and to illustrate its operation. We found that our cooperative optimization approach obtained very good solutions in significantly lower computing time.

ISBN:978-0-444-64238-7

Publication year:2018

Keywords:P1 Proceeding

BOF-keylabel:yes

Authors from:Higher Education

Accessibility:Closed