Research Paper on Coors Beer CompanyNameInstitutionThesis StatementThis article examines the case study of Coors Brewers Limited and its efforts to increase its market share through the adoption of a updated formula generated by a neural network. How effective is their adoption/what are their failures? And how can failures be remedied? Questions 1-5In order to achieve its stated goal of increasing market share, Coors must perfect a favorable product that goes beyond social stigmas, regardless of the location or event in which it is consumed. This value proposition was further complicated by the fact that Coors needed to design a product that complemented a varied potential ambiance in which it was to be consumed. Based on the market study carried out by the brewer, points of analysis and impacts were identifiable. The move was intended to increase market share through increased consumer selection relative to current market shareholders across a wide range of consumer categories. The research into securing a large market share for beer was well supported by facts and was successful. Neural networks have also helped predict beer flavor and profitability rating in areas where neural networks have been successfully applied. Neural networks provide a more general framework for relating a company's financial information to its respective bond rating. However, neural networks are not easily interpretable - the end user must be perceptive in the interpretation. The continuous process of analyzing different flavor combinations is costly in terms of cost and time. Impacts within the current process include human taste test sampling, data collection time, and costs associated with manufacturing the test product itself...... middle of paper .... .. they tackle difficult computing problems. However, based on my research on sensory evaluation models that could solve the given problem, I found one that works well. This model is known as the Multilayer Perceptron (MLP) currently selected by Coors. However, I would also recommend a submodel called Multiple Input Multiple Output (MIMO). This submodel is a specific alternative to the backpropagation design. Multiple Input, Multiple Output (MIMO) Model ReferencesHarrington, RJ (2008). Food and thread pairing: a sensory experience. Hoboken, NJ: WSiley and Sons Inc. NeuroDimension Inc. (2012). Neural network consulting. Accessed August 10, 2013 from nd.com: http://www.nd.com/resources/partners2.htmlTurban, E., Sharda, R. and Delen, D. (2011). Decision support and business intelligence systems (9th ed.). Boston: Prentice Hall..