The paper presents possible approaches for reducing the volume of data generated by simulation optimisation performed with a digital twin created in accordance with the Industry 4.0 concept. The methodology is validated using an application developed for controlling the execution of parallel simulation experiments (using client–server architecture) with the digital twin. The paper describes various pseudo-gradient, stochastic, and metaheuristic methods used for finding the global optimum without performing a complete pruning of the search space. The remote simulation optimisers reduce the volume of generated data by hashing the data. The data are sent to a remote database of simulation experiments for the digital twin for use by other simulation optimisers.

The paper deals with testing of selected heuristic optimization methods and their evaluation. We have proposed different techniques which express the success of the optimization method in different ways (the method success, the difference between optimum and local extreme, the distances of quartiles, the number of simulation experiments until the optimum was found). These evaluation techniques use box plot characteristics calculated from the repeated optimization experiments

The paper deals with testing and evaluation of selected heuristic optimization methods - Random Search, Downhill Simplex, Hill Climbing, Tabu Search, Local Search, Simulated Annealing, Evolution Strategy and Differential Evolution. We modified basic methods in such a way that they are applicable for discrete event simulation optimization purposes. The paper is mainly focused on testing Downhill Simplex and Differential Evolution because these methods achieved below-average performances in the initial testing of finding the global optimum. We modified these methods and we compared the modified and previous basic versions of these methods. We proposed different evaluation criteria (criteria express the success in different ways). These criteria use box plot characteristics calculated from the repeated optimization experiments. We have also tested different settings of these optimization methods to analyse their behaviour considering the setup of the optimization method parameters

This paper deals with the comparison of selected optimisation methods (random search, hill climbing, tabu search, local search, downhill simplex, simulated annealing, differential evolution and evolution strategy) using various evaluation criteria proposed by us which express the success of the optimisation method in different ways (the success of the method in finding the global optimum, the difference between optimum and local extremes, the distances of quartiles of possible solutions, the number of simulation experiments until the optimum was found). Some of the selected optimisation methods (downhill simplex, simulated annealing, differential evolution and evolution strategy) were modified to improve their behaviour to find the global optimum. Optimisation methods were tested on selected testing functions (De Jong's, Rosenbrock's, Michalewicz's, and Ackley's function - the substitution of the simulation model) and three discrete event simulation models (the manufacturing system and logistics, the penalty, and the assembly line). We have modified downhill simplex and differential evolution which are implemented in the simulation optimiser after the initial testing on simulation models.

The paper deals with testing and evaluation of a modified Self-Organizing Migrating Algorithm (SOMA) applied to a discrete event simulation model reflecting the supply of production lines using automated guided vehicles. The SOMA heuristic optimization method is derived from the Differential Evolution method. We test all the SOMA strategies under the same conditions of the simulation experiments – the same termination criteria, number of repetitions in the optimization experiments, and the same setting of the basic parameters of the SOMA. We propose a methodology using different evaluation criteria to analyse the different SOMA strategies behaviour of finding the optimum of an objective function specified for each discrete event simulation model.