Dynamic weight-based multi-features fuzzy fusion for recovery-decision of waste lubrication oil

Upon contaminated to a certain extent, the lubrication oil should be changed for recovery. An effective decision of the waste oil recovery process is in generally affected by factors including the vary testing indices are, contamination level and other limited conditions,called muti-features. To crack this nut, a integrated method was proposed to obtain the dynamical weights to be fused in the DS frame. Firstly, the fuzzy analytic hierarchy process (FAHP) method was proposed to solve the multi-features weights distribution by the decision makers, and the Change-weight method was used to dynamically adjust their weights by the real status, rather than the fixed weights distribution; Further, the schemes supporting information corresponding to every feature is evaluated by each decision maker, and their weights are dynamically calculated too by the joint application of technique for order preference by similarity to ideal solution (TOPSIS). The two types of dynamic weights are regarded as the basic probability assignment (BPA) to fuse the assessment information integrated by the DS theory of evidence. An example of the waste oil recovery-decision is presented to illustrate the application of the method. The effectiveness of the proposed method is validated by the example.