Using artificial neural network in demand prediction in Pichkooban company

Journal Title: Journal of Science and today’s world - Year 2015, Vol 4, Issue 7

Abstract

Prediction term base an its word meaning is imagination of a situation in future. Demand prediction is the most important issue about inventory management and for effective decision in inventory management, demand prediction for future periods are necessary. More over applying methods which results in increasing prediction precision were interested by management, economy and engineering fields. Neural networks were used as a powerful tool for predication in different fields in recent years. In this research demand prediction were considered by using neural networks. Neural network approach, multi-layered perceptron (MLP) with post-propagation learning were studied. Regarding to little information in network learning, adaptive calibration Algorithm were used. This technique has increased adaption capability of network and to some extent eliminated disadvantage of applied data loss. By using real data 1 production page type , Pichkooban company, prediction with real values compared and performance of real value methods in past period, past period mean, simple movable mean, balanced movable mean and profile smoothing with result of using.3-layer perceptron neural network were compared. In general, results show preference of MLP neural network model over other methods base on statistical performance criteria of MSE,MAD,MAPE.

Authors and Affiliations

Mohammad Mohammadi, Meisam Mohammadi

Keywords

Related Articles

Price forecasting using the non-linear model of chaos theory in Tehran Stock Exchange (Chemical Industries)

The importance of chemical industries is evident nowadays. The chemical industries in any country are considered as one of the main components of the economy. With acceptance of chemical industries into Stock Exchange...

Comparison of leisure times between youths from villages and cities of Kohgiloyeh Boyerahmad in terms of the economy status

The study aimed to compare the leisure-time situation of youths who are living in the villages and cities of Kohgiloyeh Boyerahmad, based on their economic status. The statistical population consists of young people li...

Factors Affecting the Attitude of Customers to Online Shopping (Case Study: Youth of Jahrom)

Given the widespread expansion of the Internet around the world, and the growing dependence of humans and its daily activities (such as job, shopping, leisure, communication with others, etc.) on it, it seems that the I...

Modal Analysis of Connecting Rod of Nissan Z24 Engine by Finite Element Method

Connecting rod design is very important because of its role in the crank mechanism, so that the research that can be applied in optimal design of connecting rod can lead to increased engine performance. The connecting...

Comparison of Service Quality and Customer Satisfaction in Traditional and Electronic Banking (Case Study: Parsian Bank)

Given the fact that banks are key economic factors in all countries of the world, one of the most important goals of the banks is to provide a wide range of flawless services to achieve customer satisfaction. Electroni...

Download PDF file
  • EP ID EP29559
  • DOI -
  • Views 365
  • Downloads 4

How To Cite

Mohammad Mohammadi, Meisam Mohammadi (2015). Using artificial neural network in demand prediction in Pichkooban company. Journal of Science and today’s world, 4(7), -. https://europub.co.uk/articles/-A-29559