Petroleum is an important strategic material which is connected with the vitals and safety of the national economy, and the supplier selections are related to the safety of petroleum production and supply. However, the traditional approaches for supplier selections are limited in subjective evaluation of weights, inaccurate assessing rules, and inefficient decision-making. Although most of the current methods are widely applied in corporation management, a more efficient approach needs to be proposed for supplier selection of oil enterprise.
This paper summarizes the particular characteristics of the supply chain of Chinese petroleum enterprises, analyzes the limitations of the traditional methods of supplier selection, and brought forward the method based on case reasoning system (CBR) for petroleum enterprises. The method based on data mining techniques which solves three key problems of CBR, includes calculating the weights of the attributes with information entropy in case warehouse organizing process objectively, evaluating the similarities with k-prototype clustering between the original and target cases in case retrieving process exactly, and extracting the potential rules with back propagation neural networks from conclusions in maintenance and revising process efficiently. It demonstrates the advantages, practicability and validity of this method via case study finally.
Petroleum is the most widely used energy which cannot be regenerated, and also an important strategic material that is connected to the vitals and safety of the national economy. In recent years, the supply-need contradiction of China’s energy is pricking up as Chinese economy develops rapidly. Meanwhile the strategic repertory system for oil is faulty, so China has not enough ability to deal with the oil risks. It is forecasted that in 5–15 years, the price of oil will remain in a high standard because of the factors such as the dollar depreciation, the oil production reducing, and the expectation of OPEC raising, etc.
In the existing system, the oil produced by the Chinese petroleum enterprises is purchased and sold by the government only, thus some of the oil enterprises do not have “marketing” (Zhao, 2007), therefore the suppliers of the oil enterprises take a more important role. It is quite obvious that the suppliers as the most important partners, impact the enterprises very deeply on reducing production costs, increasing the efficiency, and realizing the continuance development with their qualification, rate of technicians, high quality rate, quality of product, supply, and R&D.
Nowadays, the globalization and informationization boosted the establishment of the dynamic alliance of enterprises, and make the competitions of single enterprise into competitions of supply chains. And this changes the traditional management to the supply chain management (Ma, Lin, & Chen, 2000). Under the situation of global trade and agile supply chain, supplier selection is a very important decision for an enterprise. So how to evaluate and select the suppliers is significant for the healthy growth of an enterprise.
Because of the traditional score method for selecting suppliers is weak for dealing with the uncertain information, and the results are hard to explain, so it is difficult to adapt the new competition.
This paper will utilize the case based reasoning system which imitates human thinking abilities, combines with data mining methods to select eigenvectors with high importance, confirm the weight of attributes impersonally, evaluate the similarities of the target case and the original case accurately, and meanwhile distilling the potential rules from the selection results, thus improves the accuracy and efficiency of the decision-making of the petroleum enterprise’ supplier selection.
It is very important to select outstanding suppliers for petroleum enterprises to grow healthily under the present energy management system of China, the global business and the agile supply chain.
Case-based reasoning is a hot topic and front edge in researching in artificial intelligence and machine learning. Its obvious merits include: the complete expression of data, increment learning, the exact simulation of visualizing thinking, easy to get knowledge, and high solving efficiency, etc. But the case-based reasoning system also has a bottleneck problem of getting knowledge, such as cases, knowledge revising, similarity evaluation. Assuredly the bottleneck problem impacts the performance of CBR system. We import data mining technology into case-based reasoning to mine the data in multi data sources in CBR system, thus improved the automatization level of knowledge acquisition, performance of the system, and expedite the exploring period of the intelligent system.
Based on the analysis above, the main research goal of this paper is how to improve the accuracy and efficiency of the petroleum enterprise supplier selection decision-making, and after the theoretic research and tests of the application instances, we can get the research results below.
(1)
Based on decision-making tree and information entropy methods, we organize case warehouse with reason, distill the eigenvectors and set up their weights, thus we solved the problem that how to evaluate the importance of multiattributes objectively in the supplier decision-making.
(2)
Based on k-prototypes clustering algorithm, we take distance and module as measurement standard for numerical and text attributes, and set up a similarity evaluation model for multiattributes, solved the problem that how to evaluate multi semantic expression synthetically in supplier selection decision-making.
(3)
Based on BP neural networks method, automatically revise and maintain the case warehouse, distill rules from the decision-making results, and set up a petroleum enterprise supplier selection system based on CBR finally, and improve the efficiency and veracity of supplier selection decision-making
