In multi-stage manufacturing systems, optimisation of part quality inspection planning (PQIP) problem means to determine the optimal time, place and extent of inspection activities for assessing the significant quality characteristics of products while maximising the system efficiency. An inspection activity is capable of detecting the produced defects partially and accordingly prevents further processing of them in downstream and more importantly avoids them to reach customers. In this paper, the existing researches on the optimisation of the part quality inspection are surveyed from the viewpoint of the considered production system characteristics; the applied modelling approaches and solution methodologies. This review found that although numerous works have been already done on the PQIP, the development of multi-objective optimisation frameworks considering real production constraints under parameters uncertainty is necessary. Also, by the Industry 4.0 trend, the creation of integrated models aiming to plan the inspection, maintenance and production activities simultaneously, seems to be an important potential future research direction.
22 January 2019 in Communications in Statistics - Theory and Methods
Cross entropy is an important index for determining the divergence between two sets or distributions. Most existing cross entropy are proposed in a fuzzy environment and undefined in some uncertain situations (e.g., Dempster–Shafer theory). This study proposes an extended cross entropy measure of belief values based on a belief degree using available evidence. Thus, a new aspect of belief functions represents in the name of a belief set. Then, a new cross entropy measure between two belief sets is defined. Furthermore, the application of the cross-entropy measure in multi-criteria decision making (MCDM) is provided with belief valued information.
An increased and ongoing trend to more product variety can be observed in manufacturing industries. This trend implies that enterprises have to deal with more and more differing products which leads to challenges on both sides, the product design and the production. The improvement of production being an important factor, the challenge on the side of the design department is not minor. Between design for manufacturing/assembly and increasing demands in terms of customisation, the management of highly varying products and the generation of coherent product families is of great importance. This article presents a new design improvement methodology based on two physical and functional architecture similarity indices and two assembly technology similarity indices. It aims to guide the designer in the analysis of product similarity, the identification of product subassemblies to optimise and the choice of consistent assembly solutions for an easier product family generation and optimised production. A case study from automotive industry is presented to illustrate the application of this approach.
1 March 2023 in IAES International Journal of Artificial Intelligence (IJ-AI)
Today's world is no longer that of yesterday, the pace with which we live and also the speed is enormous and rapid, that overnight we discover the appearance of new technologies and solutions in all the fields, in particular, that of scientific research. Artificial intelligence plays the main role. Predicting the behavior of new materials using artificial neural networks has become a frequently adopted solution by researchers today. The performance of neural networks depends mainly on the activation functions used. This work was designed to mainly study the impact of these functions on the response speed of an artificial neural network in general, and particularly on the model we are working on to predict the thermomechanical behavior of innovative materials. By using TANSIG, PURELIN and LOGSIG in a feed forward back propagation training by Levenberg-Marquardt algorithm, we were able to generate 9 models. For each of these models, we were interested in analyzing the speed’s response of the network and studying its regression. Thus, this work was able to show us that choosing the right neuron activation function from one layer to another can clearly influence the performance of the results. Depending on the problem studied, the desired objective and the chosen architecture, the activation function can radically change the result and provide us with the expected efficiency.
18 December 2018 in International Journal of Production Research
This paper proposes a robust possibilistic and multi-objective mixed-integer linear programming mathematical model to concurrently plan part quality inspection and Preventive Maintenance (PM) activities for a serial multi-stage production system. This system contains the deteriorating stages and faces the uncertainty about estimated cost components and demand amount. The integrated model reaches two significant decisions which are the right time and place for performing the part quality inspection and PM. These decisions are made while the model is to simultaneously optimise the implied system productivity and total cost. To measure the implied system productivity, a new piecewise utility function for the ratio of produced conforming products to input workpieces is developed. A real case study and a numerical example are explored to validate and verify the developed model. The results prove the significance and effectiveness of considering the uncertainty and conflicting practical objectives for the problem.
2 June 2023 in International Journal of Production Research