| In recent years,with the continuous adjustment of agricultural structure and the improvement of urban and rural residents’ consumption level,the output and circulation of fruits and vegetables in China have been rising.However,COVID-19 has led to cooperation in the cold chain between fruits and vegetables,and the response speed of the cold chain has slowed down.The interruption of all links in the cold chain has caused huge losses to the participants of the chain.In the case of emergency,it is a new problem that fruit and vegetable enterprises need to solve in the post epidemic era to seek suitable cold chain partners through online platform to minimize the risk of participants in cold chain.Based on this,this paper presents a recommendation system for fruits and vegetables cold chain partners.In the uncertain environment,the hesitancy and fuzziness of the risk evaluation information of suppliers and retailers lead to the problem that the intelligent recommendation information of suppliers and retailers is difficult to express quantitatively.Therefore,a Two-sided matching model considering dual hesitant fuzzy preference information is established.The method proposed in this paper aims to recommend the lowest risk partner for multiple suppliers and retailers in the system according to the matching results in case of emergency,so as to reduce the impact of emergency on the cold chain of fruits and vegetables;provide suggestions for low risk cooperation between suppliers and retailers;strengthen the information integration and sharing,promote the cooperation of all parties in the fruit and vegetable cold chain,and promote the integrated development of primary,secondary and tertiary industries.The main work of this paper is as follows:First of all,this paper studies the risk evaluation index system of fruit and vegetable cold chain,and identifies the internal and external risk factors through the supply chain operation reference model(SCOR model)and literature analysis method.Firstly,the SCOR model and literature analysis method are selected as the research tools to identify the internal and external risks of fruit and vegetable cold chain;Then,the quantitative analysis of the identified risk factors is carried out by using the questionnaire survey method;Finally,SPSS software is used to analyze the data of the survey results,and the factors that have a greater impact on the cold chain are screened out,and the cold chain risk evaluation index system of fruits and vegetables is established.Secondly,the fruit and vegetable cold chain partner recommendation system is established.The Two-sided matching theory is used to model and analyze the partner recommendation problem in the system,and the matching algorithm is used to solve the problem.Firstly,the supplier retailer Two-sided matching problem is described,and the dual hesitant fuzzy set is used to extract the hesitant fuzzy preference information of the supplier and retailer under the evaluation index and normalize it.Secondly,considering the relationship between positive and negative ideal hesitant fuzzy elements and dual hesitant fuzzy elements in normalized dual hesitant fuzzy preference information,the bi-directional projection technique is used to calculate the bi-directional projection value of normalized hesitant fuzzy preference information of suppliers and retailers;Then,the Two-sided matching model between supplier and retailer is constructed.Finally,the fruit and vegetable cold chain partner recommendation system is established by using the model,and the system is designed and described.Thirdly,the validity and stability of the model are verified.The model is solved and verified by actual cases,and the comparative analysis and sensitivity analysis of the model are carried out.First of all,the case is described.Four suppliers and six retailers in strawberry industry are selected as the research objects,and the matching algorithm is used to solve the case,match stable cooperative retailers for every strawberry supplier. |