Research On Dynamic Two-sided Matching Method And Its Applications In Supply-demand Relationship

The problem of two-sided(bilateral)matching is the process of reasonably matching the subjects such as those between demanders and suppliers in industrial manufacturing or daily life.The traditional bilateral matching research is mostly based on single-stage static evaluation information.However,in fact,the performance of both matching parties has certain dynamic characteristics,and it is necessary to consider the dynamic evaluation information of multiple stages to make decisions.And the supply-demand relationship matching has important practical significance as a typical matching problem.Therefore,from the perspective of supply and demand relationship,this thesis studies the dynamic bilateral matching decision-making method from a multi-stage perspective.Aiming at the problem of how to obtain long-term stable matching schemes in multi-stage matching decision-making,this thesis proposes a dynamic bilateral matching method based on three-way matrix.Specifically,the model uses fuzzy grey relational analysis to convert the evaluation information of the two sides into the corresponding initial satisfactions,and then uses the combination of the initial satisfaction of the two sides to construct a three-way matrix.At the same time,the minimum satisfaction of the unilateral object is used as the standard.These combinations are divided into three different parts and candidate matching combinations are screened out.Finally,a multi-objective optimization model is established using the improved particle swarm algorithm based on the competitive swarm optimizer to obtain the matching results in a single period,and then the staged cooperation stability of the matching parties is evaluated according to the obtained results,and the above-mentioned cycle is continued to obtain the final convergent matching result.To handle the problem of bilateral matching decision based on multi-stage evaluation information,this thesis proposes a bilateral matching method based on multi-stage dynamic reference points.Because decision-makers often pay more attention to the level of gain or loss of the evaluation object relative to a reference point,and considering the dynamic changes of evaluation information in a multi-stage decision problem,this method compares the development speed of evaluation information in different stages with the average of the overall stage,and then set the reference points in different stages dynamically,this dynamic adjustment of the reference point actually played the effect of stage weighting.Then the probability density function of the interval number is used to calculate the satisfaction between the matching subjects,and a multi-objective model to maximize the satisfaction of both matching parties is constructed to obtain the matching result.The thesis carries out simulation analysis on a real case and a numerical example in the field of supply-demand relations.The comparison experimental results verify the effectiveness of the proposed methods in this thesis and the work provides an alternative way to solve the problem of supply-demand matching relationshihp.