Combining the advantages of discrete- and continuous-time scheduling models: Part 3. General algorithm

• Published in: Computers & chemical engineering Vol. 139; p. 106848
• Main Authors: Lee, Hojae, Maravelias, Christos T.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 04.08.2020
• Subjects: Chemical production scheduling; Mixed-integer programming; Process operations; Solution algorithm; Process operations; Chemical production scheduling; Solution algorithm; Mixed-integer programming
• ISSN: 0098-1354; 1873-4375
• DOI: 10.1016/j.compchemeng.2020.106848

One of the main challenges in applying optimization-based scheduling techniques in process industries stems from the different process characteristics and constraints that need to be taken into account when generating a schedule. For instance, consideration of sequence-dependent changeovers may easily make the resulting optimization model computationally expensive. Accordingly, building upon the recently proposed Discrete-Continuous Algorithm (Lee and Maravelias, 2018), we propose generalized algorithm that enables accurate and fast solution of difficult instances while efficiently handling a wide range of process characteristics. The algorithm combines modeling versatility with computational tractability while guaranteeing the feasibility and accuracy of the final solution. Through a case study inspired by a real-world brewing process, we show that our algorithm provides accurate and high quality solutions to industrial-scale instances in reasonable time.