Assessment of manufacturing lead time using neural network analysis

This paper presents a neural network (ANN) model for assessing manufacturing lead time (MLT) to increase the overall performance of a factory. MLT is affected by plant output parameters such as total number of outputs, total working hours, rework quantity, and downtime rate. These output parameters are converted to key performance indicators (KPI) to evaluate the performance of companies. These KPI's can be product efficiency, quality rate, downtime, and overall equipment efficiency. The literature shows different methods to calculate KPI's and addresses the correlation between KPI and MLT. The proposed model starts with formulas calculating the KPIs when the plant output parameters are given. Then a Neural Network model is used to build a relationship between these KPIs and the MLT. At the second stage of the paper, the neural network model is considered as an objective function to decide the best level of KPI's for minimum MLT using simulated annealing.