Neural Networks for Cost Estimation of Shell and Tube Heat Exchangers Orlando Duran * Nibaldo Rodriguez † Luiz Airton Consalter ‡ Abstract—The objective of this paper is to develop and test a model of cost estimating for the shell and tube heat exchangers in the early design phase via the application of artificial neural networks (ANN). An ANN model can help the designers to make decisions at the early phases of the design process. With an ANN model, it is possible to obtain a fairly accurate prediction, even when enough and adequate informa- tion is not available in the early stages of the design process. This model proved that neural networks are capable of reducing uncertainties related to the cost estimation of a shell and tube heat exchangers. Keywords: Cost Estimation, Heat Exchangers, Neural Networks, Shell and Tube. 1 Introduction Cost estimation is a key factor during the development phases of manufactured products. Early approximations of cost as a function of a set of general characteristics help designers in decisions such as selecting materials, produc- tion processes and mainly morphological characteristics of the product. Studies have shown that the greatest po- tential for cost reduction is at the early design phases, where as much as 80% of the cost of a product is de- cided. As the design phase itself accounts for a relatively small percentage of the total development cost, devoting a greater effort to design to cost is a reasonable and nec- essary step towards optimizing product costs. Making a wrong decision at this stage is extremely costly further down the development process. Product modifications and process alterations are more expensive the later they occur in the development cycle. Thus, cost estimators need to approximate the true cost of producing a prod- uct. In addition, since cost estimating is the start of the cost management process and influences the ’go/no-go’ decisions concerning a new product development, ideally, these go decisions regarding new product development or product design changes must be based on quantitative analysis instead of guesswork. Rush et all. [2] examine both traditional and more recent developments in cost es- * orlando.duran@ucv.cl. Pontificia Universidad Catolica de Val- paraiso, Chile. † nibaldo.rodriguez@ucv.cl, Pontificia Universidad Catolica de Valparaiso, Chile. ‡ lac@upf.br FEAR, Univesidade de Passo Fundo, Passo Fundo, RS , Brasil. timating techniques in order to highlight their advantages and limitations. The analysis includes parametric esti- mating, feature based costing, artificial intelligence, and cost management techniques. Niazi et all. [4] provide a detailed review of the state of the art in product cost es- timation covering various techniques and methodologies developed over the years. The overall work is categorized into qualitative and quantitative techniques. The qual- itative techniques are further subdivided into intuitive and analogical techniques, and the quantitative ones into parametric and analytical techniques. Curran et all. [5] provide a comprehensive literature review in engineering cost modeling as applied to aerospace. Three main quan- titative approaches can be identified for cost estimation purposes: Analogy-based techniques: these techniques are based on the definition and analysis of the degree of similarity be- tween the new product and another one, which has been already produced by the firm. The parametric method: the cost is expressed as an ana- lytical function of a set of variables that consist or repre- sent some features of the product which are supposed to influence mainly the final cost of the product. These func- tions are called Cost Estimation Relationships (CERs) and are built using statistical methodologies. Analytical Models: in this case the estimation is based on the detailed analysis of the manufacturing process and of the features of the product. The estimated cost of the product is calculated in a very analytical way, as the sum of its elementary components, constituted by the value of the resources used in each step of the production process (raw materials, components, labor, equipment, etc.). Due to this, the engineering approach can be used only when all the characteristics of the production process and of the product are well defined. Therefore, the application of this approach is limited to situation where a great amount of input data is available. Through a review in the cost estimation literature it can be observed that an incipient number of cases that use artificial intelligence (AI) techniques have been reported. These techniques constitute the last generation of tools for manufactured product cost estimation. The basic con- cept behind the application of AI in cost estimating is to Proceedings of the International MultiConference of Engineers and Computer Scientists 2008 Vol II IMECS 2008, 19-21 March, 2008, Hong Kong ISBN: 978-988-17012-1-3 IMECS 2008