A quick efficient neural network based business decision making tool in batch reactive distillation
Document Type
Article
Publication Date
1-1-2003
Abstract
Over the last decade the western world has seen marked changes in the way the manufacturing companies are operating. Many world-class bulk chemical manufacturers have significantly shrunk their businesses and or set up the businesses in the third world. This is due to increased global competition and operating cost (mainly labour cost) and strict environmental legislation. Internet has facilitated customers to choose the same products from many different companies at competitive prices in minutes rather than in days. Many UK companies no longer have the luxury to keep the same customers year after year1. The business planners have to react quickly in such a frequently changing market environment to keep the business running on profit. In this work, a quick and efficient neural network based Business Decision Making (BDM) tool is developed that can be used at all levels of the operations in manufacturing companies. This tool is especially useful at the planning level where ultimate business decision making takes place. It is demonstrated in an environment of manufacturing products using batch reactive distillation. The tool can forecast profitability, productivity, batch time, energy cost and can give optimal operating policy in few CPU seconds for changing product specifications, raw material and energy costs and products prices.
Keywords
Adaptive control, Baker's yeast, Fed-batch fermentation, Fuzzy logic, Hybrid control, Takagi-Sugeno inference method, Fuzzy control, Fuzzy sets, Substrates, Yeast, Adaptive control systems, Fuzzy logic controller, Fermentation, Saccharomyces cerevisiae.
Divisions
fac_eng
Publication Title
Computer Aided Chemical Engineering
Volume
15
Publisher
Computer Aided Chemical Engineering
Additional Information
Export Date: 5 March 2013 Source: Scopus doi: 10.1016/S1570-7946(03)80558-8 Language of Original Document: English Correspondence Address: Mujtabaa, I.M.; Engineering Modelling Group, School of Engineering, Design and Technology, University of Bradford, West Yorkshire BD7 1DP, United Kingdom; email: I.M.Mujtaba@bradford.ac.uk References: McRobbie, I., (2000) University of Bradford and A.H. Marks Ltd, Joint Research Collaboration Initiative Meeting, , 18 December, Bradford; Fogarty, D.W., Hoffmann, T.R., Stonebraker, P.W., (1989) Production and Operations Management, , South-Western Publishing Co., USA; (1997) BP Review, (October-December), pp. 15-16; Cuille, P.E., Reklaitis, G.V., (1986) Comp. Chem. Eng., 10 (4), p. 389; Mujtaba, I.M., Macchietto, S., (1997) IECRes, 36 (6), p. 2287; Sørensen, E., Skogestad, S., (1994) J. Process Control, 4 (4), p. 205; Mujtaba, I.M., Hussain, M.A., (2001) Application of Neural Network and Other Learning Technologies in Process Engineering, , Imperial College Press, USA; Greaves, M.A., Mujtaba, I.M., Barolo, M., Trotta, A., Hussain, M.A., (2002) Computer Aided Chemical Engineering, pp. 505-510. , Elsevier, London