EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

3601. An artificial neural networks model to production rate estimation of short and medium sized serial production lines

Invited abstract in session WC-6: Advancements of OR-analytics in statistics, machine learning and data science 18, stream Advancements of OR-analytics in statistics, machine learning and data science.

Wednesday, 12:30-14:00
Room: 1013 (building: 202)

Authors (first author is the speaker)

1. Mehmet Ulaş KOYUNCUOĞLU
Department of Management Information Systems, Pamukkale University

Abstract

Accurate analysis of production lines is of critical importance at strategic and tactical level for the sustainable economy of companies. Production rate, which is defined as the amount of production per unit time. The processing times, failure times and repair times (reliability parameters) of the machines and the number of buffers used between machines to reduce idle times greatly affect the production rate, which is one of the main performance indicators. Unreliable lines, where machines are subject to random failures, and balanced production lines, where the processing time of all machines is equal or expressed by independent identically distributed random variables. Exact analytical methods, approximate analytical methods and simulation are widely used for production rate calculation. Estimation of the production rates with artificial intelligence-based methods are very limited in terms of production line configuration and belong to recent years. In this study, an artificial neural network model is proposed to accurately and quickly estimate the production rate of small and medium-sized serial production lines with all machines in the line balanced, unreliable, finite buffer size. Parameter tuning is performed for the proposed artificial neural network model and additional tests are conducted to increase the prediction efficiency of the model. The results obtained encourage the proposed model to be applied to large size lines.

Keywords

Status: accepted

Back to the list of papers