Wyniki wyszukiwania - BazTech

Ograniczanie wyników

2 2025

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Adaptive Stochastic Approach for Evaluating Service Level Using a (SIM-ESL) Module Developed for Optimizing Product Availability and Order Accuracy

Aufy Samah A., Jabbar Rasha, Shiaa Waad Q.

Management Systems in Production Engineering

2025

nr 2 (33)

220-228

Inventory decisions are both high risk and high benefit throughout supply chain. Inventory management plays a crucial role since it assists manufacturing firms enhance their efficiency and competitiveness. Service level is a basic performance indicator of shortage inventory when an order is placed. Precise forecasts and optimum level of inventory can decrease the stock-outs and enhance the inventory service level. Finished goods are most common form of inventories have direct related with customer. Thus, the importance of specifying the suitable level of finished goods inventory assets for right planning and control of factory operations and optimization of the overall process to optimize production cost and time and consequently to improve service level. So, inventory management system supported by real time is requisite to ensure consistently both product availability and order accuracy. In this study, adaptive stochastic inventory management approach for evaluation service level was modelled which addressed inventory policies in order to optimize service level as a realistic value. This approach developed a Stochastic Inventory Management for Evaluating Service Level (SIMESL) module basically formulated Computational Intelligent (CI) in such a way that evaluated Availability Of Product (AOP) and Accuracy Of Order (AOO). It is structured with two models are: Evaluation Dynamic Maximum Economic Order Quantity (EDEOQ) model developed Artificial Neural Network (ANN) to evaluate the optimum inventory level of finished goods as a function of product varied demand and lead time. The second one, Variability Between-Among Dynamic Economic Order Quantity (VB-ADEOQ) is modeled a fuzzy logic to estimate the performance measures (AOP and AOO) that make sure the required order is being processed and delivered on time. Finally, the effectively of the adaptive approach is evaluated through a tested theoretical data. The obtained results were proof it’s efficiently in exactly evaluating of an unavailable the required demand about 46.937% and delivered to their final destination without errors about 56.655%. While, Taguchi method used to optimize and determine relative magnitude of the effect of various factors and their interactive. The obtained results were proved and reliability in determining the customer satisfaction score in real-world. Finally, the developed module was proved their reliability and accurately in addressing the vague data, uncertainty, no formulas, and varied inputs in determining the customer satisfaction score in real-world. Consequently, it clearly makes the developed module a better choice for inventory level forecasting, and then it will assist manufacturing industries improve their business process procedures without incurring significant capital expenditure.

Implement the artificial neural network concept for predicting the mechanical properties of printed polylactic acid parts

Al-Wswasi Mazin, Al-Khaleeli Weam A., Aufy Samah A.

Advances in Science and Technology. Research Journal

2025

Vol. 19, no. 5

73--83

Additive Manufacturing (AM) is an industrial process that involves creating three-dimensional (3D) parts based on computer-aided design (CAD) models. Various methods and techniques have been developed in the recent decade to enhance this industry. This research observes the influence of 3D printing parameters using fused deposition modeling (FDM) on the uniaxial compressive strength (UCS) of polylactic acid (PLA) specimens. This is precisely to study the effects of infill density, infill pattern, and layer thickness and determine the optimal parameters. The compression test samples have been designed based on ASTM D695 standards and manufactured using a Creality Ender-5 Pro 3D printer. Then, a Taguchi design of experiments method has been used, and nine experiments have been conducted to evaluate the effects of the mentioned parameters. Also, analysis of variance (ANOVA) declared that the infill density is the most noticeable parameter with a contribution of 83.56% to the variation in UCS. On the other hands, both infill pattern and layer thickness had minimal impact. However, the ideal configuration to earn maximum UCS value has been recorded as 80% infill density, a gyroid infill pattern, and a 0.3 mm layer thickness based on ANOVA analysis. Furthermore, an artificial neural network (ANN) model has been developed to enhance predictive capabilities. This is by training a three-layer architecture with inputs of infill density, infill pattern, and layer thickness. It is confirmed by two calculation outcomes that the ANN has performed high predictive accuracy: a regression coefficient (R) of 0.9974 and slight deviation between experimental and predicted UCS values. These results show the considerable role of infill density in increasing the compressive strength, as well as approve the ANN as a trusted tool for predicting mechanical properties of 3D-printed components. This research presents profound investigation for optimizing FDM parameters to enhance the mechanical performance of 3D-printed parts.