Search the digital library catalog

Abstract: Reporting the empirical results of swarm and evolutionary computation algorithms is a challenging task with many possible difficulties. These difficulties stem from the stochastic nature of such algorithms, as well as their inability to guarantee an optimal solution in polynomial time. This research deals with measuring the performance of stochastic optimization algorithms, as well as the confidence intervals of the empirically obtained statistics. Traditionally, the arithmetic mean is used for measuring average performance, but we propose quantiles for measuring average, peak and bad-case performance, and give their interpretations in a relevant context for measuring the performance of the metaheuristics. In order to investigate the differences between arithmetic mean and quantiles, and to confirm possible benefits, we conducted experiments with 7 stochastic algorithms and 20 unconstrained continuous variable optimization problems. The experiments showed that median was a better measure of average performance than arithmetic mean, based on the observed solution quality. Out of 20 problem instances, a discrepancy between the arithmetic mean and median happened in 6 instances, out of which 5 were resolved in favor of median and 1 instance remained unresolved as a near tie. The arithmetic mean was completely inadequate for measuring average performance based on the observed number of function evaluations, while the 0.5 quantile (median) was suitable for that task. The quantiles also showed to be adequate for assessing peak performance and bad-case performance. In this paper, we also proposed a bootstrap method to calculate the confidence intervals of the probability of the empirically obtained quantiles. Considering the many advantages of using quantiles, including the ability to calculate probabilities of success in the case of multiple executions of the algorithm and the practically useful method of calculating confidence intervals, we recommend quantiles as the standard measure of peak, average and bad-case performance of stochastic optimization algorithms.
Keywords: algorithmic performance, experimental evaluation, metaheuristics, quantile, confidence interval, stochastic algorithms, evolutionary computation, swarm intelligence, experimental methodology
Published in DKUM: 28.03.2025; Views: 0; Downloads: 9
Full text (490,48 KB)
This document has many files! More...

Abstract: To meet the rising demands of global trade and e-commerce, efficient warehousing relies on inte-grated and cooperative material handling systems. This paper investigates the extension of VerticalLift Module (VLM) storage capability with a Buffer System and assesses the impact of this integra-tion on performance. We developed an analytical model to calculate the expected dual commandcycles, forming the basis for evaluating the VLM – Buffer integrated storage system’s performance.Our research emphasises minimising unnecessary swaps between the VLM and the Buffer Systemto enhance throughput performance. We introduce the Look Ahead Strategy (LAS) to minimiseinter-system swaps and develop a Binary Integer Program (BIP) to benchmark its performance. Theresults indicate that LAS performs on par with BIP, due to its ability to consider product popular-ity during the final selection of the outbound swapping tote. Through a comprehensive analysisof the analytical model with an empirical correction, utilising Pareto-based order sequences, theresults show deviations of less than 1% on average, affirming the analytical model’s accuracy. Ourresearch provides insights on using the VLM-Buffer integrated storage system, emphasising efficienttote swapping policies like LAS for enhanced warehouse operations, and allows managers to assesssystem performance through scenario-based analyses.
Keywords: automated warehouses, vertical lift module, toteswapping, analytical and numerical modelling, performance evaluation
Published in DKUM: 11.10.2024; Views: 0; Downloads: 30
Full text (3,56 MB)
This document has many files! More...

Abstract: Supplier evaluation and selection is essential to any organization, and planning an effective and comprehensive approach to that end seems inevitable. Meanwhile, determining the requisite criteria for evaluating and selecting suppliers is probably one of the most important steps to be taken towards developing an evaluation and selection model in the organization. In this article, first a review of the literature on the criteria and the field of supplier evaluation and selection are provided. These criteria are then placed into proper categories. In order to formulate a supplier evaluation and selection framework for the manufacturing organization under study, the implemented categorization is applied where a list of fifteen attributes and performance criteria is created; where upon it is secured with the help of a designated panel (project team). These features are then screened using Lawshe's method the "social attribute" is removed from the list of fifteen. The remaining 14 other criteria are configured within the SEAP (Suppliers Evaluation based on Attributes and Performances) framework. The framework follows the objective of continually evaluating suppliers, both potential and actual ones through incorporating their performances into their qualification ratings. Based on the proposed framework, suppliers are evaluated on the basis of two types of criteria, - feature (attribute) and performance.
Keywords: framework, supplier evaluation, supplier selection, criteria, performance measurement, attributes, Lawshe's method
Published in DKUM: 22.08.2024; Views: 73; Downloads: 8
Full text (617,06 KB)
This document has many files! More...

Abstract: Robotic bin-picking is increasingly important in the order-picking process in intralogistics. However, many aspects of the robotic bin-picking process (object detection, grasping, manipulation) still require the research community's attention. Established methods are used to test robotic grippers, enabling comparability of the research community's results. This study presents a modified YCB Robotic Gripper Assessment Protocol that was used to evaluate the performance of four robotic grippers (twofingered, vacuum, gecko, and soft gripper). During the testing, 45 objects from the modified YCB Object and Model Set from the packaging categories, tools, small objects, spherical objects, and deformable objects were grasped and manipulated. The results of the robotic gripper evaluation show that while some robotic grippers performed substantially well, there is an expressive grasp success variation over diverse objects. The results indicate that selecting the object grasp point next to selecting the most suitable robotic gripper is critical in successful object grasping. Therefore, we propose grasp point determination using mechanical software simulation with a model of a two-fingered gripper in an ADAMS/MATLAB cosimulation. Performing software simulations for this task can save time and give comparable results to real-world experiments.
Keywords: intralogistics, robotic bin-picking, YCB protocol, robotic gripper evaluation, mechanical software simulations, performance analysis
Published in DKUM: 21.03.2024; Views: 298; Downloads: 20
Full text (773,38 KB)
This document has many files! More...