Kansei engineering-based artificial neural network model to evaluate worker performance in small-medium scale food production system

This paper highlighted a new method to evaluate worker performance in small medium-scale food production system. By using Kansei engineering, worker performance can be analysed using verbal parameter of profile of mood states and non-verbal parameter of heart rate in a given workplace environment. Fusing various parameters of worker performance requires a robust modelling tool. An artificial neural network (ANN) model is proposed to evaluate worker performance based on categories of normal, capacity constrained and over capacity workers. The training and inspection data were recapitulated from four types of food production systems as tempe, bakpia, fish chips and cracker. The ANN was trained using back-propagation supervised learning method and inspection data. The trained ANN models produced satisfied correlation between measured and predicted value and minimum inspection error. The research result is applicable not only for building Kansei engineering-based sensor, but also for decision support for production planning and control in food production system.