Aristotle University Of Thessaloniki (AUTH)
Optimization-based techniques for production scheduling of continuous and batch processes
Abstract
dc:descriptionDuring the last few years, the significant advances of cutting-edge technologies led to a fourth industrial revolution, referred to as Industry 4.0. At the dawn of the new era of industrialization, modern production environments attempt to integrate various decisional and physical aspects of production processes into automated and decentralized systems. Two of the main decision levels in these systems, are production planning and scheduling, which constitute a major component for the efficient operation of the process industries. Especially in the current competitive globalized market, production planning and scheduling are of vital importance to most industries, since profit margins are miniscule. Therefore, efficient usage of resources has a critical role in the viability and sustainability of all industries. Additionally, efficiency targets are increasingly being adapted with sustainable production goals towards a green and circular economy. In addition to cost savings, further objectives must be considered, such as the reduction of greenhouse gas emissions, the increased usage of renewable energy sources and the reduction of waste. These objectives can be achieved by exploiting recent advances of computer-aided optimization tools and methodologies. During the last 30 years plethora of research contributions have been published by the scientific community in the field of production scheduling optimization. However, the practical implementation of optimization-based scheduling frameworks in real-life industrial applications is limited. In most industries, the optimization of production scheduling constitutes an extremely challenging and time-consuming task, since the majority of decision-makers prefer to generate scheduling solutions manually, or use simulation-based software, resulting to suboptimal solutions. This thesis proposes systematic mathematical frameworks for the optimization of a wide variety of complex production planning and scheduling problems. The optimization-based solutions are based on mixed integer linear programming (MILP) frameworks. However, a main drawback of MILP models is their inability to handle efficiently large problem instances, since the model size increases exponentially with the problem size. To face this challenge, novel MILP-based solution algorithms have been also investigated for the solution of real-life industrial problems.More specifically, the first chapter considers the scheduling problem of a real-life large-scale industrial facility of packaged consumer goods. The problem under consideration is mainly focused on the packing stage which constitutes the major production bottleneck. Two precedence-based MILP mathematical models are proposed to describe explicitly the continuous process of the plant. The models rely on allocation, timing and sequencing constraints. Additional constraints, referring to the production/formulation stage of the plant, are also imposed in order to ensure the generation of feasible production schedules. Furthermore, two MILP-based decomposition algorithms are proposed for the efficient solution of large-scale problem instances. The applicability of the proposed approaches is illustrated by solving several real-life industrial problem instances of a multinational consumer goods industry under consideration. The results lead to nearly optimal scheduling in reasonable solution times, comparing favorably with manually derived schedules by the production engineers.The second chapter addresses the scheduling problem of continuous make-and-pack industries, including flexible intermediate storage vessels, aiming to provide better synchronisation of the production stages. A novel continuous-time, precedence-based, MILP model is developed for the problem under consideration. Extending previously proposed precedence-based MILP models, multiple campaigns of the same recipe can be stored simultaneously in a storage tank. Explicit resource constraints related to the generation and recycling of byproduct are introduced, to achieve a better utilization of the available resources. Several case studies, inspired by a large-scale consumer goods industry have been solved, to illustrate the applicability of the proposed frameworks. Although global optimal solutions cannot be guaranteed, good quality schedules are obtained, while the utilisation of intermediate buffers leads to a better synchronisation of the production stages and increased productivity.The final chapter of the thesis presents an integrated planning and scheduling framework for the optimal contract selection problem of Contract Manufacturing Organizations (CMOs) under uncertainty in pharmaceutical industry. During the last 20 years a growing number of pharmaceutical companies outsource part of their operations to reduce operational cost and mitigate their risk exposure. Contract Manufacturing Organizations (CMOs) utilize their facilities to manufacture products for multinational pharmaceutical companies on a contract basis. Considering a multistage, multiproduct, batch facility of a secondary pharmaceutical industry, an aggregated MILP planning model, including material balances and allocation constraints is firstly proposed. Using a rolling horizon approach, the production targets are then provided to a precedence-based MILP scheduling model to define batch-sizing and sequencing decisions in detail. To model demand uncertainty, a scenario-based approach is proposed, considering the Value-at-Risk (VaR) and Conditional Value-at-Risk (CVaR) measures. Since large number of scenarios introduces significant challenges to computations, a scenario reduction framework is integrated to reduce the total solution time, when considering large-scale problem instances. The proposed methodology increases the profitability of CMOs, by selecting the optimal contract combinations, depending on their risk tolerance, while considering the availability and optimal utilization of underlying production resources.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Elekidis, Apostolos
- Ελεκίδης, Απόστολος
Subjects
dc:subject × 15- Βελτιστοποίηση
- Χρονοπρογραμματισμός παραγωγής
- Βιομηχανικές διεργασίες
- Διαχείριση ρίσκου
- Μαθηματική μοντελοποίηση
- Mathematical modelling
- Optimization
- Process industry
- Risk management
- Production scheduling
- Επιστήμη Χημικού Μηχανικού
- Επιστήμες Μηχανικού και Τεχνολογία
- Χημική μηχανική
- Chemical Engineering
- Engineering and Technology
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/51999