Increased Productivity andSustainability with DigitalProduction Optimization from NRW
The OptiProd.NRW team consists of experienced ICT and simulation experts, leading researchers in the field of optimal production schedulung, and global players in the process industry. In the following, the partners introduce themselves.
INOSIM Software GmbH (Coordinator), Dortmund, Germany
INOSIM Software GmbH was founded in 2003 with the goal of providing the chemical, pharmaceutical, biotechnological, and related industries with powerful process simulation software. Today, INOSIM is a global technology leader in this field, counting practically all global players as well as many other, mostly highly specialized technology companies, among its customers. INOSIM software is used for a variety of tasks, e. g., for simulation-supported process design, the review, evaluation and optimization of production and logistics concepts, and simulation-driven analysis, optimization and production scheduling for existing production plants. Furthermore, INOSIM is cooperating with several leading companies in the process industry to establish its simulation software also for simulation-based decision support in modern batch production.
INOSIM Software GmbH is an experienced partner in R&D projects, both on the national and on the European level. The INOSIM team has extensive experience in the analysis and support of manual optimization of production plans on the basis of simulation with detailed material flow models. In numerous consulting projects for the design of new production plants as well as for the analysis and optimization of existing plants, INOSIM has achieved considerable success in all branches of the process industry, even without the use of mathematical optimization, e. g. increasing the production volume by 30% or more through debottlenecking and changes in production schedules alone, or avoiding plant extensions by increasing the capacity utilization of production plants. Notwithstanding these successes, it is expected that further significant savings and efficiency improvements can be achieved by mathematical optimization. Modern production plants are extremely complex, which means that only a very small selection of possible production plans can be created and checked manually. Moreover, mathematical optimizers are also innovation generators, since, unlike humans, they also check unusual and counterintuitive candidates which may offer potential for improvement that human personnel would never find.
INOSIM employee Dr. -Ing. Christian Sonntag, who takes on the role of coordinator in the project, has longstanding experience in the acquisition, management and technological implementation of national and EU projects (FP6, FP7, H2020). These include CONSENS, HYCON, HYCON2, MULTIFORM, DYMASOS, LEGOLAS, and SkaMPi. He is also an expert in application-driven research and development, including modeling and simulation, mathematical optimization, and optimal control. Together with Prof. S. Engell (partner TU Dortmund), he has worked on strategy development for various technological areas, including cyber-physical systems (CPS), the Internet of Things (IoT), Industry 4. 0, and Systems of Systems (SoS) within the EU projects CPSoS (coordinated by Prof. Engell) and PICASSO. Partner INOSIM coordinates the project and contributes to all technical work packages. The focus of the work is to develop an integrative software system which brings together all components developed in the project. To this end, INOSIM provides its technical and industrial expertise as well as its extensive experience in the management of R&D projects to the project.
Process Dynamics And Operations Group (DYN), TU Dortmund University, Germany
The Process Dynamics And Operations Group (DYN) at TU Dortmund University, which is led by Prof. Dr. -Ing. Sebastian Engell, is one of the world's leading research institutes in the field of optimal operation of production plants. The main focus of its work is on model-predictive control, the plant-wide monitoring and optimization of the efficiency of coupled production plants, and the calculation of optimal production schedules in batch processing.
The DYN group has been working on methods of production scheduling and control in batch processing for a long time, yielding e.g. new mathematical optimization formulations as well as "unconventional" methods such as timed-automata-based optimization and the use of evolutionary algorithms. The work of the DYN group on the optimization of production schedules in so-called Pipeless Plants are of particular relevance for OptiProd.NRW. These have a large number of interacting degrees of freedom, ranging from the selection of processing stations for specific operations over sequencing to route planning for driverless transport systems. In the OptiProd.NRW project, an approach is being pursued which is based on the combination of meta-heuristics with underlying scheduling mechanisms. It is suitable for complex tasks with a high number of degrees of freedom.
Bayer AG, Leverkusen, Germany
The Bayer AG from Leverkusen is a life science company with a history of over 150 years and core competences in the areas of health care and agriculture. Bayer employs numerous experts who regularly use process simulation software to model, analyze and structurally improve production plants. The Bayer experts will provide their extensinve experience in plant design and the design of detailed production schedules to the project. They have done simulation projects on plants of a variety of tyes, including chemical and biological ingredient production, product formulation and packaging, and the related logistics. This knowledge will enable the systematic modelling and testing of a suitable case study of the project in cooperation with the other project partners. Bayer's expert knowledge also plays a central role in the evaluation of the project results. The company has been cooperating intensively with TU Dortmund university and INOSIM in research and development for many years.
In particular, Bayer has been using simulation software from INOSIM Software GmbH for many years to analyze and improve its production facilities. In some cases, these models are used at the production sites to check the feasibility of production plans generated by the ERP planning system and to implement improvements that were determined through simulation. These projects have shown that the level of detail and hence the predictive quality of the INOSIM material flow models is superior to traditional planning systems. The models contain complex decision rules that are applied to assign free resources (units, raw materials, personnel, etc.) to production orders automatically. While this approach always provides feasible detailed production schedules, it requires extensive manual presorting and prioritization of the order lists to achieve usable performance. Optimal, or even reactive real-time, production scheduling is therefore not yet current practice at Bayer. In cooperation with TU Dortmund university, Bayer is therefore also working on mathematical optimization approaches already. Due to the complexity and the enormous number of degrees of freedom in production, these approaches have so far only been able to find optimal detailed production schedules for medium-sized systems.
Bayer has also developed a methodology for using material flow models in life cycle analysis. In this context, resource and energy efficiency as well as corresponding improvement measures can be systematically evaluated. This approach provides very reliable results due to its holistic model approach.
