Multi-criteria optimization strategy for production chains of the graph structure

  Project Title:

Multi-criteria optimization strategy for production chains of the graph structure
Principal Investigator:

Prof. dr hab. inż. Jan Kusiak

 Project ID:

2013/11/B/ST8/00352

 Sources of funding:

NCN (OPUS)
Acronim:

StratOptim

 Project duration (years):

2014/09/01 – 2017/02/28

 Key words:

optimization strategy, multi-criteria optimization, optimization of production chains
Description of the results of the project:
Production chains that can be observed in industrial plants inspired research undertaken in the project. They usually consist of numerous physical stands where semi-products are created. Then quality of these semi-products is assessed, and when sufficiently high, they are transferred to further processing in subsequent stand. Optimal solutions at each stage, may or may not lead to optimal quality of the final product.

 

In the project we elaborated and analyzed several optimization strategies, their effectiveness and utility in reaching high quality of the final product. At start we focused on theoretical models using several benchmark functions and selected most promising methods. These methods were next applied to the real metallurgical multi-stage problems:

 

· Copper blister production (copper of high purity) according to the Kennecott–Outokumpu technology,

· Refining and production of lead in Kaldo convertor,

· Hydrometallurgical process of zinc production.

 

The mentioned above processes were modeled by standard regression tools and artificial intelligence methodologies on the basis of historical data describing these processes. In theoretical models, success of the methods was highly correlated with the choice of benchmark functions. In all considered technological problems mentioned above sequential approach, based on the knowledge of the structure of described process was most successful. Among other things, it shows that in metallurgical processes, where success of one stage depends highly on quality achieved in previous stages, sequential approach has an advantage. Optimization methodologies developed in the project can be applied to any production chains of linear structure with possible branching.
List of the publications: 1. P. Jarosz, J. Kusiak, S. Małecki, P. Oprocha, Ł. Sztangret, M. Wilkus, A methodology for optimization in multistage industrial processes: a pilot study, Mathematical Problems in Engineering, 2015, s. 1–10.

2. M. Gulik, P. Jarosz, J. Kusiak, S. Małecki, P. Morkisz, P. Oprocha, W. Pietrucha, Ł. Sztangret, Metamodelling and optimization of copper flash smelting process, Proceedings of the 3rd international conference on Optimization techniques in engineering (OTENG’15): Rome, Italy, November 7–9, 2015 / ed. Imre J. Rudas. — [Italy]: WSEAS Press, cop. 2015. s. 52–56.

3. P. Jarosz, J. Kusiak, S. Małecki, P. Morkisz, P. Oprocha, W. Pietrucha, Ł. Sztangret, Metamodelowanie procesu produkcji miedzi blister – porównanie efektywności metod modelowania, Materiały konferencji Informatyka w technologii metali (KomPlasTech 2016): Wisła, 17–20 stycznia 2016,s. 70–71.

4. M. Gulik, P. Jarosz, J. Kusiak, S. Małecki, P. Oprocha, Ł. Sztangret, Modelowanie procesu produkcji miedzi blister z wykorzystaniem sztucznych sieci neuronowych, Rudy i Metale Nieżelazne Recykling 61 nr 1, 2016, s. 21–25.

5. P. Jarosz, J. Kusiak, S. Małecki, P. Morkisz, P. Oprocha, W. Pietrucha, Ł. Sztangret, Metamodeling and optimization of a blister copper two-stage production process, JOM / The Minerals, Metals & Materials Society vol. 68 no. 6, 2016, s. 1535–1540.

6. J. Kusiak, P. Morkisz, P. Oprocha, W. Pietrucha, Ł. Sztangret, On aggregation of stages in multi-criteria optimization of chain structured processes, Lecture Notes in Artificial Intelligence, Switzerland: Springer International Publishing, 2016, s. 411–419.

7. J. Kusiak, P. Morkisz, P. Oprocha, W. Pietrucha, Ł. Sztangret, Multi-criteria optimization strategies for production chains, Proceedings of the12th international conference on Numerical Methods in Industrial Forming processes (NUMIFORM 2016), Troyes, France, July 4–7, 2016, s.1-6.

8. P. Jarosz, J. Kusiak, S. Małecki, P. Morkisz, P. Oprocha, W. Pietrucha, Ł. Sztangret, Evaluation of optimization strategies dedicated to multistage processes, Proceedings of the 24th international conference on Computer Methods in Materials Technology (KomPlasTech 2017): Zakopane, Poland, January 15–18, 2017, s. 171-174.

9. J. Kusiak, P. Morkisz, P. Oprocha, W. Pietrucha, Ł. Sztangret, Validation of optimization strategies using the linear structured production chains, AIP Conference Proceedings of the 2017 International Conference on Applied Mathematics and Computer Science (ICAMCS 2017), Rome, Italy, January 27‐29, 2017 (in print).

10. Ł. Sztangret, J. Kusiak, An attempt of reduction of optimization costs of complex industrial processes, AIP Conference Proceedings of the 2017 International Conference on Mathematical Methods & Computational Techniques in Science & Engineering, University of Cambridge, UK, February 24-26, 2017 (in print).

11. P. Jarosz, J. Kusiak, S. Małecki, P. Morkisz, P. Oprocha, W. Pietrucha, Ł. Sztangret, Multi-criteria optimization strategies for production chains with branching, International Journal of Material Forming (in review).

12. P. Jarosz, J. Kusiak, S. Małecki, P. Morkisz, P. Oprocha, W. Pietrucha, Ł. Sztangret, An Attempt of Optimization of Zinc Production Line, Metals and Materials International (in review).