Disjunctive multiperiod optimization models for process networks

Multiperiod optimization models for design and planning in the chemical industry have received considerable attention in recent years. In this paper, we present a general multiperiod Mixed Integer Nonlinear Programming (MINLP) model, which incorporates design, as well as operation and expansion planning, and takes into account the investment costs, operating costs and expansion costs in each time period. We present a hybrid algorithm for solution of the model and compare the solution time to that of DICOPT++, a commercial solver. The proposed algorithm makes use of disjunctive programming, as well as a bilevel decomposition technique proposed by Iyer and Grossmann (1997). A numerical example is presented and results show that the proposed method shows a significant decrease in total solution time compared to DICOPT++. Results also show that problems with a significant number of time periods could be solved in reasonable time.