Multiperiod design and planning of multiproduct batch plants with mixed‐product campaigns

Abstract

This work presents a multiperiod optimization model for multiproduct batch plants operating during several time periods with different characteristics because of seasonal and market fluctuations. This model simultaneously considers decisions about the design, operation, scheduling, and planning of the plant and the corresponding tradeoffs among them. Thus, decomposition mechanisms, which have been frequently used in previous approaches, are avoided through a formulation that takes into account the main elements of these problems. Besides, decisions are affected by different context conditions arisen by the multiperiod effect. Through a mixed integer nonlinear program, different alternatives of mixed production campaign are considered, handled by means of a novel set of scheduling constraints. This approach is posed for a fermentors network with high detail level in the description of the unit operations in a plant that produces yeast and ethanol. V

A multiproduct batch plant design model incorporating production planning and scheduling decisions under a multiperiod scenario

“…Given the important combinatorial characteristic of the problem, the proposed approach involved coupling a stochastic algorithm, specifically a genetic algorithm, with a discrete-event simulator. Working with similar plants, Corsano et al [15] developed a multiperiod formulation in order to optimize design and production planning simultaneously, where they employed MPCs to solve the production scheduling. Using a mixed-integer non linear programming (MINLP) formulation, a set of possible production campaigns were employed, which were handled by predetermined scheduling constraints, where this approach was applied to a fermentation network.…”

Optimal design of pre-fermentation and fermentation stages applying nonlinear programming

“…Grisi et al [20] presented a MILP formulation for the short-term scheduling of the integrated processes for sugar, bioethanol, biogas and bioelectricity production. In Corsano et al [21] a MINLP model for the simultaneous optimization of the design, operation, scheduling, and planning of a fermentation network is proposed. The optimal production campaign considering a multiperiod approach is obtained.…”