Neuroevolutionary Metamodeling and Optimization: inventory control

Project Type: Academic

Main Challenge:

Taking into consideration the urgent industrial need in metamodeling automation and recent success of neuroevolutionary approaches in neural architecture search and hyperparameter optimization, this aims to examine feasibility and efficiency of the combination of artificial neural network and genetic algorithm for automated metamodeling of inventory control systems. The developed framework is built upon the most prominent state-of-the-art practices and demonstrates solid computational capabilities in classical metamodeling formulated as a regression problem. Additionally, the possibility of using the proposed framework to derive optimal control parameters is discussed with regard to a real-world case study.

Funding: VIAA scholarship for doctoral research, decision N.1.-50.3/3889, decision N.1.-50.3/2978

DA&AI tools:

• Artificial Neural Networks
• Neural Architecture Search
• Genetic algorithm

Deliverables (publications, reports, acknowledgments, etc.)

• Jackson, I. (2020) GitHub repository “metainventory” [online] Available at: https://github.com/Jackil1993/metainventory
• Jackson, I. and Tolujevs, J. (2018) The Discrete-Event Approach to Simulate Stochastic Multi-Product (Q, r) Inventory Control Systems. In: proceedings of the 28th International Conference on Information Modelling and Knowledge Bases (EJC-2018), June 4-8, 2018, Riga, Latvia. pp. 94-101 [indexed by Scopus].
• Jackson, I., Tolujevs, J. and Reggelin, T. (2018) The Combination of Discrete-Event Simulation and Genetic Algorithm for Solving the Stochastic Multi-Product Inventory Optimization Problem. Transport and Telecommunication Journal, 19(3), pp. 233-243 [indexed by Scopus, indexed by Web of Science].
• Jackson, I., Tolujevs, J., Lang, S. and Kegenbekov, Z. (2019) Metamodeling of Inventory-Control Simulations Based on a Multilayer Perceptron. Transport and Telecommunication Journal, 20(3), pp. 251-259 [indexed by Scopus, indexed by Web of Science].
• Jackson, I. and Tolujevs, J. (2018) A combination of simulation and genetic algorithm for solving a stochastic inventory optimization problem. In: proceedings of 11th International Doctoral Student Workshop on Logistics, June 19, 2018, Magdeburg, Germany, pp. 31-35.
• Jackson, I. and Tolujevs, J. (2018) Simulation-driven artificial intelligence for solving stochastic combinatorial optimization problems in production and logistics. In: proceedings of PhD seminar Sci-Bi: Digitalization in Logistics and Transport, Riga, Latvia, pp. 30-36
• Jackson, I. and Tolujevs, J. (2019) Metamodeling of Inventory-Control Systems Based on Artificial Neural Networks. In: proceedings of 12th International Doctoral Student Workshop on Logistics, June 6, 2019, Magdeburg, Germany, pp. 82-86.
• Jackson, I. (2019) Simulation-Optimization Approach to Stochastic Inventory Control with Perishability. Information Technology & Management Science, 22, pp. 9-14.
• Jackson, I. (2019) Combining Simulation with Genetic Algorithm for Solving Stochastic Multi-Product Inventory Optimization Problem. The Journal of Economic Research & Business Administration, 130(4), pp. 96-102.
• Jackson, I. (2019) Neuroevolutionary Approach to Metamodeling of Production-Inventory Systems with Lost-Sales and Markovian Demand. In: proceedings of International Conference on Reliability and Statistics in Transportation and Communication. Springer, Cham, pp. 90-99 [indexed by Scopus].
• Jackson, I., Tolujevs, J. and Kegenbekov, Z. (2020) Review of Inventory Control Models: A Classification Based on Methods of Obtaining Optimal Control Parameters. Transport and Telecommunication Journal, 21(3), pp. 191-202 [Scopus, Web of Science].