Supplier selection method, supplier selection system and supplier selection device

Abstract

The present invention provides a supplier selection method, a supplier selection system and a supplier selection device, which include the following steps: determining the subjective and objective indicators of suppliers; inviting industry experts to score the influence relationship between the subjective and objective indicators of suppliers, and establishing an initial subjective and objective indicator direct relationship matrix; standardizing the initial subjective and objective indicator direct relationship matrix; calculating the subjective and objective indicator comprehensive influence matrix based on the standardized subjective and objective indicator direct relationship matrix; determining the centrality and causality of the subjective and objective indicators, and obtaining the subjective and objective indicator weights; determining the decision value of the subjective and objective indicators, and obtaining the loss coefficient of the subjective and objective indicators; obtaining the subjective and objective loss value of the supplier; calculating the total loss value of the supplier, and ranking the suppliers to determine the optimal supplier. The present invention achieves the technical effect of enterprises' rapid and accurate selection management of suppliers.

Description

Technical Field

[0001] The present invention relates to the field of supplier selection, and in particular to a supplier selection method, a supplier selection system and a supplier selection device. Background Art

[0002] The supply of fresh agricultural products is closely related to people's lives and production, and has long been the focus of the country and society. Supermarkets, catering companies, agricultural product processing companies, etc. are all faced with the problem of selecting suppliers of fresh agricultural products. The purchase of fresh agricultural products is the starting point and key link in the supply chain of fresh agricultural products, which determines the safety, quality and price of the products. At this stage, the procurement methods are becoming more and more market-oriented and diversified. If fresh agricultural product retailers want to expand the market, gain user trust and achieve sustainable development, it is particularly important to choose suppliers with high credibility, safety and sustainability. At this stage, when evaluating suppliers of fresh agricultural products, most of them are based on subjective impressions and the purchasing experience of buyers. It is easy to fall into subjective traps and there are great risks. Summary of the invention

[0003] In view of the deficiencies in the prior art, the present invention provides a supplier selection method, a supplier selection system and a supplier selection device based on DEMATEL and Taguchi quality loss function, which can accurately evaluate each fresh agricultural product supplier. Enterprises do not need to contact fresh agricultural product suppliers one by one, which reduces the communication cost of enterprises and improves communication efficiency. The technical effect of achieving rapid and accurate selection and management of fresh agricultural product suppliers by enterprises is achieved.

[0004] The present invention achieves the above technical objectives through the following technical means.

[0005] A supplier selection method based on DEMATEL and Taguchi quality loss function includes the following steps:

[0006] Determine the subjective and objective indicators of suppliers;

[0007] Invite industry experts to score the impact relationship between suppliers' subjective indicators, take the arithmetic mean of the impact relationship scores between subjective indicators, and establish an initial direct relationship matrix of subjective indicators; Invite industry experts to score the impact relationship between suppliers' objective indicators, take the arithmetic mean of the impact relationship scores between objective indicators, and establish an initial direct relationship matrix of objective indicators;

[0008] Standardize the initial subjective indicator direct relationship matrix and the initial objective indicator direct relationship matrix respectively;

[0009] The subjective indicator comprehensive influence matrix is ​​calculated based on the standardized direct relationship matrix of subjective indicators; the objective indicator comprehensive influence matrix is ​​calculated based on the standardized direct relationship matrix of objective indicators;

[0010] Determine the centrality and causality of the subjective index to obtain the weight of the subjective index; determine the centrality and causality of the objective index to obtain the weight of the objective index;

[0011] Based on the Taguchi quality loss function method, the decision values ​​of subjective and objective indicators are determined, and the loss coefficients of subjective and objective indicators are obtained;

[0012] Obtain the supplier's subjective loss value and objective loss value;

[0013] Calculate the total loss value of suppliers, sort the suppliers and determine the best supplier.

[0014] Furthermore, the initial subjective indicator direct relationship matrix M is established 1 And the initial objective indicator direct relationship matrix M 2 ,as follows:

[0015]

[0016] Where:

[0017] m is the number of subjective indicators; n is the number of objective indicators;

[0018] a ij It means the arithmetic mean of the influence relationship between the i-th subjective indicator and the j-th subjective indicator. ij In M 1 represents the value of the i-th row and j-th column; when i=j, a ij =0; a' gh It means the arithmetic mean of the influence relationship between the gth objective indicator and the hth objective indicator, a' gh In M 2 represents the value of the gth row and hth column; when g=h, a' gh =0.

[0019] Furthermore, the calculation formulas for the standardized initial subjective indicator direct relationship matrix and the initial objective indicator direct relationship matrix are as follows:

[0020] X k =λ k *M k (k=1,2),

[0021]

[0022] In the formula: k = 1 represents subjective indicators, k = 2 represents objective indicators;

[0023] λ 1 is the standardized coefficient of the subjective index, λ 2 is the standardized coefficient of the objective indicator;

[0024] X 1 is the standardized direct relationship matrix of subjective indicators;

[0025] X 2 It is the standardized direct relationship matrix of subjective indicators.

[0026] Furthermore, the calculation formula for the comprehensive impact matrix of subjective / objective indicators is:

[0027] T k =X k (EX k ) -1

[0028] Where:

[0029] E is the identity matrix;

[0030] Subjective indicator comprehensive impact matrix T 1 It is expressed as:

[0031] Objective indicator comprehensive impact matrix T 2 It is expressed as:

[0032] in:

[0033] t ij Represents the matrix T 1 The comprehensive influence of the i-th subjective indicator on the j-th subjective indicator, t' gh Represents the matrix T 2 The comprehensive influence of the g-th objective indicator on the h-th objective indicator.

[0034] Furthermore, the centrality and causality of the subjective / objective indicators are determined as follows:

[0035] Let f k,z is the matrix T k The sum of each row of k,l is the matrix T k The sum of each column of ;

[0036] When k = 1, z = i, l = j, the formula is as follows:

[0037]

[0038] When k = 2, z = g, l = h, the formula is as follows:

[0039]

[0040]

[0041] Determine the centrality and causality of subjective / objective indicators:

[0042] u k,z =f k,z +e k,l ,

[0043] v k,z =f k,z -e k,l ,

[0044] Where:

[0045] When k = 1, u 1,i is the centrality of the ith subjective indicator; v 1,i is the cause degree of the i-th subjective indicator;

[0046] When k = 2, u 2,g is the centrality of the g-th objective indicator; v 2,g is the cause degree of the g-th objective indicator.

[0047] Furthermore, the subjective / objective indicator weights are obtained as follows:

[0048]

[0049] Where:

[0050] w 1,i is the weight of the i-th subjective indicator; w 2,g is the weight of the g-th objective indicator.

[0051] Furthermore, based on the Taguchi quality loss function method, the decision values ​​of subjective and objective indicators are determined, and the loss coefficients of subjective and objective indicators are obtained, specifically:

[0052] According to the supplier's historical data, the subjective index and the objective index are divided into the expected large characteristic index and the expected small characteristic index, and the maximum value of the expected large characteristic index and the index limit of the expected large characteristic index are determined according to the supplier's historical data; the maximum value of the expected small characteristic index and the index limit of the expected small characteristic index are determined according to the supplier's historical data;

[0053] According to the maximum value of the Wangda characteristic index and the index limit of the Wangda characteristic index, the loss coefficient of the Wangda characteristic index is determined by the quality loss function of the Wangda characteristic index, where: the quality loss function of the Wangda characteristic index is: L(y) b =K b / y2 , K b The loss coefficient of the expected large characteristic index, y represents the index limit of the expected large characteristic index; L(y) b Represents the maximum value of the Wangda characteristic index;

[0054] According to the maximum value of the Wang Xiao characteristic index and the index limit of the Wang Xiao characteristic index, the loss coefficient of the Wang Xiao characteristic index is determined by the quality loss function of the Wang Xiao characteristic index, where: the quality loss function of the Wang Xiao characteristic index is: L(y) s =K s *y′ 2 , K s The loss coefficient of the expected small characteristic index, y′ represents the index limit of the expected small characteristic index; L(y) s Represents the maximum value of the small characteristic index.

[0055] Furthermore, the supplier's subjective loss value and objective loss value are obtained, specifically:

[0056] According to the supplier's historical data and the loss coefficient of the Wangda characteristic index, the loss value of the Wangda characteristic index is determined by using the quality loss function of the Wangda characteristic index; according to the supplier's historical data and the loss coefficient of the Wangxiao characteristic index, the loss value of the Wangxiao characteristic index is determined by using the quality loss function of the Wangxiao characteristic index;

[0057] According to the loss value of the large characteristic index and the loss value of the small characteristic index, the corresponding subjective index loss value L is obtained. 1,i And the objective indicator loss value L 2,g ; L 1,i represents the loss value of the i-th subjective indicator, L 2,g represents the loss value of the g-th objective indicator;

[0058] Calculate the total loss value of suppliers and sort the suppliers to determine the optimal sequence, specifically:

[0059] Calculating subjective loss value Calculate objective loss value

[0060] Set a subjective loss value w based on the supplier's historical data 1 and the objective loss value w 2 ;

[0061] The total loss value of the supplier is L = w 1 *L 1 +w 2 *L 2 ;

[0062] Sort the loss value of each supplier from large to small and determine the optimal supplier.

[0063] A supplier selection system comprising:

[0064] Database unit: used to store, manage and update fresh agricultural product supplier information;

[0065] Procurement project release unit: used to release products that need to be purchased;

[0066] Information acquisition unit: used to obtain subjective and objective indicator information of suppliers;

[0067] Evaluation and selection unit: Based on the subjective and objective indicator information of suppliers obtained by the information acquisition unit, the supplier selection method based on DEMATEL and Taguchi quality loss function is used to score the suppliers, and the suppliers are sorted and selected based on the scores.

[0068] A supplier selection device, comprising:

[0069] Memory: used to store the program written by the supplier selection method based on DEMATEL and Taguchi quality loss function;

[0070] Processor: used to process and execute programs.

[0071] The beneficial effects of the present invention are:

[0072] The supplier selection method based on DEMATEL and Taguchi quality loss function, the supplier selection system and the supplier selection device described in the present invention can accurately evaluate each supplier through the proposed sustainable fresh agricultural product supplier selection method based on DEMATEL and Taguchi quality loss function, and then screen / select suppliers according to the evaluation results of each supplier, thereby achieving the technical effect of the enterprise accurately and quickly selecting suppliers; on the other hand, the fresh agricultural product supplier selection management device proposed in the present invention has a built-in fresh agricultural product supplier selection method, which can effectively store, manage and update the information of fresh agricultural product suppliers and procurement projects, and at the same time greatly improves the efficiency of the enterprise in selecting fresh agricultural product suppliers. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. The drawings described below are some embodiments of the present invention. For ordinary technicians in this field, it is obvious that other drawings can be obtained based on these drawings without paying any creative work.

[0074] Figure 1 The present invention provides a flow chart of the supplier selection method based on DEMATEL and Taguchi quality loss function.

[0075] Figure 2 It is a schematic diagram of an application scenario of a supplier selection device in an embodiment of the present invention. DETAILED DESCRIPTION

[0076] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments, but the protection scope of the present invention is not limited thereto.

[0077] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0078] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "axial", "radial", "vertical", "horizontal", "inner", "outer" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0079] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0080] This embodiment applies the supplier selection method based on DEMATEL and Taguchi quality loss function described in the present invention to the fresh agricultural product supply chain. Retailers preliminarily screen qualified fresh agricultural product suppliers through the fresh agricultural product supplier database module, and then select the best supplier based on the score of each supplier. The advantage of this is that retailers only need to input the subjective and objective indicator information of the supplier into the information processing module, and obtain the best choice through the built-in algorithm of the information processing module.

[0081] Assume that a fresh produce retailer needs to select a suitable fresh produce supplier for a certain type of fresh produce. There are four alternative fresh produce suppliers (A 1 ,A 2 ,A 3 ,A 4 ).

[0082] S1: Determine the subjective and objective indicators of suppliers. There are 3 subjective indicators and 4 objective indicators in the evaluation indicators of fresh agricultural product suppliers. The subjective indicators are: reliability, 1 , Maintainability 2 And product appearance 3 , objective indicators include: price ob 1 , lead time 2 , sampling inspection pass rate 3 and sold out rate 4 . 3 evaluation experts, namely ex 1 、ex 2 and ex 3 The influence relationship between the subjective indicators of fresh agricultural product suppliers is scored separately.

[0083] Reliability 1 and maintainability 2 The index aims to reflect the sustainability and stable supply capacity of fresh agricultural product suppliers. Each fresh agricultural product supplier is scored based on the experience of multiple experts (purchasing specialists), and the arithmetic average is taken to obtain the final score;

[0084] Product appearance 3 The indicators are difficult to describe quantitatively. Fresh agricultural product suppliers need to provide samples before the purchase demand occurs. Multiple experts (purchasing specialists) score the samples and take the arithmetic average to get the final score. When the products are accepted later, the scores of the previous samples will be used to check whether the purchased products are qualified.

[0085] Price indicator ob 1 It means that after the purchase demand is issued by the fresh agricultural product supplier, each fresh agricultural product supplier will make a corresponding product quotation;

[0086] Lead time 2 It refers to the total time from when a purchase order is placed to when the product is put into storage. The length of the lead time is particularly important for fresh agricultural product retailers to quickly seize the market.

[0087] Sampling inspection pass rate 3It refers to the sampling inspection of fresh agricultural product retailers after receiving the products. The inspection items include product appearance, freshness, taste, pesticide residues, etc. The pass rate of random inspection can be expressed by the following formula:

[0088]

[0089] In the formula, T represents the number of products sampled, and t represents the number of qualified products sampled;

[0090] Sold out rate 4 It refers to the ratio of the sales volume of the fresh agricultural product supplier's products in the previous round of sales activities to the purchase volume, which reflects the popularity of the products provided by the fresh agricultural product supplier from the side. The expression is as follows:

[0091]

[0092] In the formula, N represents the quantity of product purchased, and q represents the quantity of product sold;

[0093] The selection of the above indicators is only an example of an embodiment of the present invention. Different enterprises can reasonably select indicators according to their business content and characteristics.

[0094] During the evaluation process, the evaluation experts used the fuzzy language variables in Table 1 to score the influence relationship between subjective and objective indicators. Taking three subjective variables as an example, the scoring results of the mutual influence relationship between subjective indicators by three evaluation experts are shown in Table 2; the scores of the three experts were combined and the arithmetic mean was taken to obtain the initial direct relationship matrix of subjective indicators, as shown in Table 3.

[0095] Table 1 Fuzzy language variable table

[0096] Influence Relationship Value range No impact 0 Weak impact (0，0.2) Weak impact (0.2，0.5) Strong influence (0.5，0.8) Significant impact (0.8，1)

[0097] Table 2 Scoring of the mutual influence relationship between subjective indicators by three experts

[0098]

[0099] Table 3 Initial direct relationship matrix of subjective indicators

[0100]

[0101]

[0102] S2: The calculation formula for the standardized initial subjective indicator direct relationship matrix and the initial objective indicator direct relationship matrix is:

[0103] X k =λ k *M k (k=1,2),

[0104]

[0105] In the formula: k = 1 represents subjective indicators, k = 2 represents objective indicators;

[0106] λ 1 is the standardized coefficient of the subjective index, λ 2 is the standardized coefficient of the objective indicator;

[0107] X 1 is the standardized direct relationship matrix of subjective indicators;

[0108] X 2 It is the standardized direct relationship matrix of subjective indicators.

[0109] The data in Table 3 are standardized to obtain the standardized subjective index direct relationship matrix X 1 , as shown in Table 4:

[0110] Table 4 Standardized subjective indicator direct relationship matrix

[0111] <![CDATA[X 1 ]]> <![CDATA[su 1 ]]> <![CDATA[su 2 ]]> <![CDATA[su 3 ]]> <![CDATA[su 1 ]]> 0 0.84 0.16 <![CDATA[su 2 ]]> 0.84 0 0.16 <![CDATA[su 3 ]]> 0.08 0.08 0

[0112] S3: The calculation formula for the comprehensive impact matrix of subjective indicators is: T k =X k (EX k ) -1 , where: E is the unit matrix; subjective index comprehensive influence matrix T 1 It is expressed as:

[0113] Objective indicator comprehensive impact matrix T 2 It is expressed as:

[0114] in:

[0115] t ij Represents the matrix T 1 The comprehensive influence of the i-th subjective indicator on the j-th subjective indicator, t' gh Represents the matrix T 2 The comprehensive influence of the g-th objective indicator on the h-th objective indicator.

[0116] Obtain the comprehensive influence matrix T of subjective indicators 1 , as shown in Table 5:

[0117] Table 5 Comprehensive influence matrix of subjective indicators

[0118] <![CDATA[T 1 ]]> <![CDATA[su 1 ]]> <![CDATA[su 2 ]]> <![CDATA[su 3 ]]> <![CDATA[su 1 ]]> 2.992 3.4485 1.1905 <![CDATA[su 2 ]]> 3.4485 2.992 1.1905 <![CDATA[su 3 ]]> 0.5952 0.5952 0.1905

[0119] S4: Determine the centrality and causality of the subjective indicator to obtain the weight of the subjective indicator; determine the centrality and causality of the objective indicator to obtain the weight of the objective indicator, specifically:

[0120] Determine the centrality and causality of the subjective / objective indicators as follows:

[0121] Let f k,z is the matrix T k The sum of each row of k,l is the matrix T k The sum of each column of ;

[0122] When k = 1, z = i, l = j, the formula is as follows:

[0123]

[0124] When k = 2, z = g, l = h, the formula is as follows:

[0125]

[0126] Determine the centrality and causality of subjective / objective indicators:

[0127] u k,z =f k,z +e k,l ,

[0128] v k,z =f k,z -e k,l ,

[0129] Where:

[0130] When k = 1, u 1,i is the centrality of the ith subjective indicator; v 1,i is the cause degree of the i-th subjective indicator;

[0131] When k = 2, u 2,g is the centrality of the g-th objective indicator; v 2,g is the cause degree of the g-th objective indicator.

[0132] Determine the weights of subjective / objective indicators as follows:

[0133]

[0134] Where:

[0135] w 1,i is the weight of the i-th subjective indicator; w 2,g is the weight of the g-th objective indicator.

[0136] The weights of the three subjective indicators calculated in the embodiment are: reliability su 1 The weight is 0.4384, maintainability su 2 The weight of 0.4384 and product appearance su 3 The weight is 0.1232.

[0137] Use the same steps to calculate the weights of each indicator in the objective indicator: price ob 1 The weight is 0.2666, the lead time ob 2 The weight is 0.2400, and the sampling pass rate is ob 3 The weight is 0.2074 and the sold-out rate is ob 4 The weight is 0.2860.

[0138] S5: Based on the Taguchi quality loss function method, the decision values ​​of subjective and objective indicators are determined, and the loss coefficients of subjective and objective indicators are obtained, specifically:

[0139] According to the supplier's historical data, the subjective indicators and objective indicators are divided into large characteristic indicators and small characteristic indicators. The maximum value of the large characteristic indicator and the indicator limit of the large characteristic indicator are determined according to the supplier's historical data; the maximum value of the small characteristic indicator and the indicator limit of the small characteristic indicator are determined according to the supplier's historical data; the meaning of the large characteristic indicator is that the larger the value, the better the indicator, and the meaning of the small characteristic indicator is that the smaller the value, the better the indicator.

[0140] According to the maximum value of the Wangda characteristic index and the index limit of the Wangda characteristic index, the loss coefficient of the Wangda characteristic index is determined by the quality loss function of the Wangda characteristic index, where: the quality loss function of the Wangda characteristic index is: L(y) b =K b / y 2 , K b The loss coefficient of the large characteristic index is expected. When calculating the loss coefficient of the large characteristic index, substitute the index limit of the large characteristic index into y and the maximum value of the large characteristic index into L(y). b , the loss coefficient of the expected characteristic index is calculated, and the calculated values ​​in the embodiment are shown in Table 6.

[0141] Table 6 Target values ​​and index limits of Wangda characteristic indicators

[0142] index Weight Target value (%) Index Limit (%) scope(%) <![CDATA[Loss coefficient (K b )]]> <![CDATA[su 1 ]]> 0.44 100 80 80-100 64 <![CDATA[su 2 ]]> 0.44 100 75 75-100 56.25 <![CDATA[su 3 ]]> 0.12 100 80 80-100 64 <![CDATA[ob 3 ]]> 0.21 100 95 95-100 90.25 <![CDATA[ob 4 ]]> 0.28 100 80 80-100 64

[0143] In the above table, the indicator su 1 For example, L(y) b =K b / y 2 ; L b (su 1)=100(the index set by the retailer 1 can obtain the maximum value), y = 0.8 (the indicator su set by the retailer 1 The lower limit of K b (su 1 )=100*(0.8) 2 =64.

[0144] According to the maximum value of the Wang Xiao characteristic index and the index limit of the Wang Xiao characteristic index, the loss coefficient of the Wang Xiao characteristic index is determined by the quality loss function of the Wang Xiao characteristic index, where: the quality loss function of the Wang Xiao characteristic index is: L(y′) s =K s *y′ 2 , K s The loss coefficient of the expected small characteristic index is calculated. When calculating the loss coefficient of the expected large characteristic index, the index limit of the expected small characteristic index is substituted into y′, and the maximum value of the expected small characteristic index is substituted into L(y′). b , the loss coefficient of the expected small characteristic index is calculated, and the calculated values ​​in the embodiment are shown in Table 7.

[0145] Table 7 Target values ​​and index limits of the small-scale characteristic index

[0146] index Weight Target value (%) Index Limit (%) scope(%) <![CDATA[Loss coefficient (K s )]]> <![CDATA[ob 1 ]]> 0.27 0 90 0-90 123.46 <![CDATA[ob 2 ]]> 0.24 0 10 0-10 10000

[0147] In the above table, the indicator ob 1 For example, L(y′) s =K s *y′ 2 ; L s (ob 1 )=100(the indicator ob set by the retailer 1 can obtain the maximum value), y′=0.9 (the indicator ob set by the retailer 1 The upper limit of K s (ob 1 )=100 / (0.9) 2 =123.46.

[0148] S6: Obtain the supplier's subjective loss value and objective loss value, specifically:

[0149] According to the supplier's historical data and the loss coefficient of the Wangda characteristic index, the loss value of the Wangda characteristic index is determined by using the quality loss function of the Wangda characteristic index; according to the supplier's historical data and the loss coefficient of the Wangxiao characteristic index, the loss value of the Wangxiao characteristic index is determined by using the quality loss function of the Wangxiao characteristic index;

[0150] Select 4 fresh agricultural product suppliers (A 1 ,A 2,A 3 ,A 4 ) Based on the historical data in the retailer database, we obtained the score table of each indicator of each fresh agricultural product supplier, as shown in Table 8.

[0151] Table 8 Rating table of various indicators for fresh agricultural product suppliers

[0152] supplier <![CDATA[su 1 ]]> <![CDATA[su 2 ]]> <![CDATA[su 3 ]]> <![CDATA[ob 1 ]]> <![CDATA[ob 2 ]]> <![CDATA[ob 3 ]]> <![CDATA[ob 4 ]]> <![CDATA[A 1 ]]> 86 81 87 7 3 96 86 <![CDATA[A 2 ]]> 80 76 90 4 5 99 90 <![CDATA[A 3 ]]> 82 79 92 6 2 97 92 <![CDATA[A 4 ]]> 87 80 91 9 3 99 87

[0153] Calculate the subjective and objective indicator loss values ​​of each supplier, and take supplier A as the most valuable supplier in the Wangda characteristic indicator. 1 The indicator 1 For example, L b (su 1 )=K b (su 1 ) / 0.86 2 =64 / 0.86 2 =86.5333; In the small characteristic index, supplier A 1 The indicator ob 1 For example, L s (ob 1 )=K s (ob 1 )*0.07 2 =123.46*0.07 2 =0.6050. The loss values ​​of each indicator of each fresh agricultural product supplier are obtained, as shown in Table 9.

[0154] Table 9 Loss value of each indicator of each fresh agricultural product supplier

[0155] supplier <![CDATA[su 1 ]]> <![CDATA[su 2 ]]> <![CDATA[su 3 ]]> <![CDATA[ob 1 ]]> <![CDATA[ob 2 ]]> <![CDATA[ob 3 ]]> <![CDATA[ob 4 ]]> <![CDATA[A 1 ]]> 86.5333 85.7339 84.5554 0.6050 9.0000 97.9275 86.5333 <![CDATA[A 2 ]]> 100.0000 97.3857 79.0123 0.1975 25.0000 92.0824 79.0123 <![CDATA[A 3 ]]> 95.1814 90.1298 75.6144 0.4445 4.0000 95.9188 75.6144 <![CDATA[A 4 ]]> 84.5554 87.8906 77.2854 1.0000 9.0000 92.0824 84.5554

[0156] S7: Calculate the total loss value of suppliers and sort the suppliers to determine the optimal sequence, specifically:

[0157] Calculating subjective loss value Calculate objective loss value

[0158] Set a subjective loss value w based on the supplier's historical data 1 and the objective loss value w 2 ;

[0159] The total loss value of the supplier is L = w 1 *L 1 +w 2 *L 2 ;

[0160] In the embodiment, the loss value of each indicator of each fresh agricultural product supplier in Table 9 is multiplied by the weight of the corresponding subjective / objective indicator to obtain the weighted loss value of each indicator of each supplier, as shown in Table 10:

[0161] Table 10 Weighted loss values ​​of various indicators of each supplier

[0162] supplier <![CDATA[su 1 ]]> <![CDATA[su 2 ]]> <![CDATA[su 3 ]]> <![CDATA[ob 1 ]]> <![CDATA[ob 2 ]]> <![CDATA[ob 3 ]]> <![CDATA[ob 4 ]]> <![CDATA[A 1 ]]> 37.9362 37.5857 10.4172 0.1613 2.1600 20.3102 24.7485 <![CDATA[A 2 ]]> 43.8400 42.6939 9.7343 0.0527 6.0000 19.0979 22.5975 <![CDATA[A 3 ]]> 41.7275 39.5129 9.3157 0.1185 0.9600 19.8936 21.6257 <![CDATA[A 4 ]]> 37.0691 38.5313 9.5216 0.2666 2.1600 19.0979 24.1829

[0163] The subjective loss value of each supplier’s subjective indicator is added together to obtain the supplier’s subjective loss value, and the objective loss value of each supplier’s objective indicator is added together to obtain the supplier’s objective loss value, as shown in Table 11.

[0164] Table 11 Subjective and objective loss values ​​of various fresh agricultural product suppliers

[0165] supplier Subjective loss value Objective loss value <![CDATA[A 1 ]]> 85.9391 47.3800 <![CDATA[A 2 ]]> 96.2682 47.7481 <![CDATA[A 3 ]]> 90.5561 42.5978 <![CDATA[A 4 ]]> 85.1219 45.7074

[0166] The weights of the subjective loss value and the objective loss value are given as w 1 =0.3, w 2 =0.7, and the total loss value of each fresh agricultural product supplier is calculated, as shown in Table 12.

[0167] Table 12 Total loss value of each fresh agricultural product supplier

[0168] supplier <![CDATA[A 1 ]]> <![CDATA[A 2 ]]> <![CDATA[A 3 ]]> <![CDATA[A 4 <!-- 11 -->]]> Total loss value 58.9477 62.3041 56.9853 57.5317

[0169] According to the calculation results, the four fresh agricultural product suppliers are ranked according to their total loss value: A 3 <A 4 <A 1 <A 2 Therefore, among the four alternatives, fresh produce supplier A 3 Best for this fresh produce retailer.

[0170] An embodiment of the present invention further provides a supplier selection system, the system comprising:

[0171] Database unit: used to store, manage and update fresh agricultural product supplier information.

[0172] Procurement project release unit: used to release products to be purchased, including product name, category, specification, quantity and other procurement requirements;

[0173] Information acquisition unit: obtains subjective and objective indicator information of fresh agricultural product suppliers, and at the same time obtains existing indicator information of fresh agricultural product suppliers in previous purchase orders from the database unit, all of which are used for the review and selection unit;

[0174] Evaluation and selection unit: Based on the subjective and objective indicator information of fresh agricultural product suppliers obtained by the information acquisition module, the fresh agricultural product suppliers are scored using the DEMATEL and Taguchi quality loss function selection method proposed above, and they are ranked and selected based on the loss scores.

[0175] After the purchase order is placed, fresh agricultural product suppliers of the same category are preliminarily screened out based on the category of the purchased fresh agricultural products, and fresh agricultural product suppliers that do not meet the requirements are excluded based on previous transaction data.

[0176] The subjective indicators of fresh agricultural product suppliers are scored by multiple experts based on their own professional skills, and the final subjective score is obtained by taking the arithmetic mean; the objective indicators of fresh agricultural product suppliers can all be obtained through quantitative description, and the obtained subjective and objective indicator information is entered into the supplier interactive information interface.

[0177] The supplier review and selection unit calculates the final loss score of each fresh agricultural product supplier based on the information on the supplier interactive information interface, and selects the appropriate fresh agricultural product supplier based on the loss score.

[0178] The embodiment of the present invention further provides a supplier selection device, including:

[0179] Memory: used to store the program written by the supplier selection method based on DEMATEL and Taguchi quality loss function;

[0180] Processor: used to process programs;

[0181] The program refers to the sustainable fresh agricultural product supplier selection method based on DEMATEL and Taguchi quality loss function proposed by the present invention, and the method is executed by a processor.

[0182] The memory stores a program, and the memory includes: a floppy disk, an optical disk, a DVD, a hard disk, a flash memory, a USB disk, a CF card, an SD card, an MMC card, a SM card, a memory stick, and other media that can store program codes.

[0183] like Figure 2 As shown, the supplier selection system mentioned above can be applied to smart terminals such as laptops, desktops, tablets, and smart phones in the form of software or clients.

[0184] It should be understood that although this specification is described according to various embodiments, not every embodiment contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

[0185] The series of detailed descriptions listed above are only specific descriptions of feasible embodiments of the present invention. They are not intended to limit the scope of protection of the present invention. All equivalent embodiments or changes that do not deviate from the technical spirit of the present invention should be included in the scope of protection of the present invention.

Claims

1. A supplier selection method, characterized in that: The steps include: Determine the subjective and objective indicators of suppliers; subjective indicators include reliability, maintainability and product appearance; objective indicators include price, lead time, sampling pass rate and sell-out rate; Invite industry experts to score the impact relationship between suppliers' subjective indicators, take the arithmetic mean of the impact relationship scores between subjective indicators, and establish an initial direct relationship matrix of subjective indicators; Invite industry experts to score the impact relationship between suppliers' objective indicators, take the arithmetic mean of the impact relationship scores between objective indicators, and establish an initial direct relationship matrix of objective indicators; Standardize the initial subjective indicator direct relationship matrix and the initial objective indicator direct relationship matrix respectively; The subjective indicator comprehensive influence matrix is ​​calculated based on the standardized direct relationship matrix of subjective indicators; the objective indicator comprehensive influence matrix is ​​calculated based on the standardized direct relationship matrix of objective indicators; Determine the centrality and causality of the subjective indicator to obtain the subjective indicator weight; determine the centrality and causality of the objective indicator to obtain the objective indicator weight, as follows: Let f k,z is the matrix T k The sum of each row of k,l is the matrix T k The sum of each column of ; When k = 1, z = i, l = j, the formula is as follows: When k = 2, z = g, l = h, the formula is as follows: Determine the centrality and causality of subjective / objective indicators: u k,z =f k,z +e k,l , v k,z =f k,z -e k,l , Where: When k = 1, u 1,i is the centrality of the ith subjective indicator; v 1,i is the cause degree of the i-th subjective indicator; T1 is the comprehensive influence matrix of subjective indicators; m is the number of subjective indicators; When k = 2, u 2,g is the centrality of the g-th objective indicator; v 2,g is the cause degree of the g-th objective indicator; T2 is the comprehensive influence matrix of objective indicators; n is the number of objective indicators; The subjective / objective indicator weights are as follows: Where: w 1,i is the weight of the i-th subjective indicator; w 2,g is the weight of the g-th objective indicator; Based on the Taguchi quality loss function method, the decision values ​​of subjective and objective indicators are determined, and the loss coefficients of subjective and objective indicators are obtained, specifically: According to the supplier's historical data, the subjective index and the objective index are divided into the expected large characteristic index and the expected small characteristic index, and the maximum value of the expected large characteristic index and the index limit of the expected large characteristic index are determined according to the supplier's historical data; the maximum value of the expected small characteristic index and the index limit of the expected small characteristic index are determined according to the supplier's historical data; According to the maximum value of the Wangda characteristic index and the index limit of the Wangda characteristic index, the loss coefficient of the Wangda characteristic index is determined by the quality loss function of the Wangda characteristic index, where: the quality loss function of the Wangda characteristic index is: L(y) b =K b / y 2 , K b The loss coefficient of the expected large characteristic index, y represents the index limit of the expected large characteristic index; L(y) b Represents the maximum value of the Wangda characteristic index; According to the maximum value of the Wang Xiao characteristic index and the index limit of the Wang Xiao characteristic index, the loss coefficient of the Wang Xiao characteristic index is determined by the quality loss function of the Wang Xiao characteristic index, where: the quality loss function of the Wang Xiao characteristic index is: L(y′) s =K s *y′ 2 , K s The loss coefficient of the expected small characteristic index, y′ represents the index limit of the expected small characteristic index; L(y′) s Represents the maximum value of the small characteristic index; The supplier's subjective loss value and objective loss value are obtained as follows: According to the supplier's historical data and the loss coefficient of the Wangda characteristic index, the loss value of the Wangda characteristic index is determined by using the quality loss function of the Wangda characteristic index; according to the supplier's historical data and the loss coefficient of the Wangxiao characteristic index, the loss value of the Wangxiao characteristic index is determined by using the quality loss function of the Wangxiao characteristic index; According to the loss value of the large characteristic index and the loss value of the small characteristic index, the corresponding subjective index loss value L is obtained. 1,i And the objective indicator loss value L 2,g ; L 1,i represents the loss value of the i-th subjective indicator, L 2,g represents the loss value of the g-th objective indicator; Calculate the total loss value of suppliers, sort the suppliers, and determine the best supplier, specifically: Calculating subjective loss value Calculate objective loss value Set the subjective loss value w1 and the objective loss value w2 based on the supplier’s historical data; The total loss value of the supplier L = w1*L1+w2*L2; Sort the loss value of each supplier from large to small and determine the optimal supplier.

2. The supplier selection method according to claim 1, characterized in that: The initial subjective indicator direct relationship matrix M1 and the initial objective indicator direct relationship matrix M2 are established as follows: Where: m is the number of subjective indicators; n is the number of objective indicators; a ij It means the arithmetic mean of the influence relationship between the i-th subjective indicator and the j-th subjective indicator. ij In M1, it represents the value of row i and column j; When i=j, a ij =0; a' gh It means the arithmetic mean of the influence relationship between the gth objective indicator and the hth objective indicator, a' gh In M2, it represents the value of the gth row and the hth column; when g = h, a' gh =0.

3. The supplier selection method according to claim 2, characterized in that: The calculation formula of the standardized initial subjective indicator direct relationship matrix and the initial objective indicator direct relationship matrix is: X k =λ k *M k ,k=1,2, In the formula: k = 1 represents subjective indicators, k = 2 represents objective indicators; λ1 is the standardized coefficient of the subjective indicator, and λ2 is the standardized coefficient of the objective indicator; X1 is the standardized direct relationship matrix of subjective indicators; X2 is the standardized direct relationship matrix of subjective indicators.

4. The supplier selection method according to claim 3, characterized in that: The calculation formula for calculating the comprehensive impact matrix of subjective / objective indicators is: T k =X k (E-X k ) -1 Where: E is the identity matrix; The comprehensive influence matrix T1 of subjective indicators is expressed as: The comprehensive impact matrix T2 of objective indicators is expressed as: in: t ij represents the comprehensive influence of the i-th subjective indicator in the matrix T1 on the j-th subjective indicator, t' gh It represents the comprehensive influence of the g-th objective indicator in the matrix T2 on the h-th objective indicator.

5. A supplier selection system, characterized in that: include: Database unit: used to store, manage and update fresh agricultural product supplier information; Procurement project release unit: used to release products that need to be purchased; Information acquisition unit: used to obtain subjective and objective indicator information of suppliers; Evaluation and selection unit: Based on the subjective and objective indicator information of suppliers obtained by the information acquisition unit, the supplier selection method according to claim 1 is used to score the suppliers, and the suppliers are sorted and selected based on the scores.

6. A supplier selection device, characterized in that: include: Memory: used to store a program written by the supplier selection method according to claim 1; Processor: used to process and execute programs.

Images