CN102542520A - Supplier cluster analysis management and customer allocation method - Google Patents

Abstract

The invention discloses a supplier cluster analysis management and customer allocation method. Firstly, the supplier is clustered and analyzed, and then the customer is allocated according to the analysis result, including: using the attribute-oriented induction algorithm, according to the first selected induction attribute and the second selected inductive attribute, cluster analysis is performed on all suppliers; the control system sets the evaluation rules according to the description; suppliers whose grade value in the supplier category described by the first category is lower than the predetermined lower threshold All the suppliers in the category are extracted, and the level value of the corresponding second category description belongs to it, and it is judged whether the level value is still lower than the predetermined lower threshold; The supplier profile data whose grade value is still lower than the predetermined lower threshold value described by the second category is put into the backup supplier database. Through the method of the invention, suppliers can be selected fairly.

Description

A supplier cluster analysis management and customer allocation method

technical field

The invention relates to a supplier management and control method, in particular to a supplier cluster analysis management and customer allocation method.

Background technique

In the field of mobile communication service provider operations, with the explosive growth of service content, reasonable management and control of suppliers has increasingly become an essential means of modern business. Common operator control needs to go through many links to complete, and human subjective intervention is strong, resulting in substantial unfairness.

Contents of the invention

Aiming at the drawbacks of the prior art, the present invention provides a supplier cluster analysis management and customer allocation method, which can use a computer system to automatically analyze the status of suppliers and allocate customer resources to achieve efficient and fair management and allocation Way.

The present invention provides a supplier cluster analysis management and customer allocation method, which is characterized in that, firstly, the supplier is clustered and analyzed, and then the customer is allocated according to the analysis result,

Specifically include the following steps:

1) Using the attribute-oriented induction algorithm, according to the first selected induction attribute, perform cluster analysis on all suppliers, and generate the first type description of each type of supplier after clustering;

2) Using the attribute-oriented inductive algorithm again, according to the second selected inductive attribute, perform cluster analysis on all suppliers, and generate the second type description of each type of supplier after clustering;

3) The control system sets the first evaluation rule according to the description of the first type, sets the second evaluation rule according to the description of the second type, and respectively evaluates all supplier types according to the first evaluation rule and the second evaluation rule. carry out ratings;

4) Extract all the suppliers in the supplier category with the first category description whose level value is lower than the predetermined lower threshold, retrieve the level value of the corresponding second category description, and judge its level value Whether it is still below the predetermined lower threshold;

5) Put the supplier profile data whose grade value is lower than the predetermined lower threshold described by the first category and whose grade value is still lower than the predetermined lower threshold described by the second category into the standby supplier database, and put the remaining suppliers Put the supplier profile data into the available supplier database;

6) Excavate the data and data of each parameter in the available suppliers, discover the correlation degree between each parameter, and evaluate the correlation degree level of each parameter;

7) Use the distribution processing center to receive customer information, decompose the customer information into demand information and contact information, encrypt the demand information and contact information separately, and generate their own security keys;

8) Excavate the data in the demand information, match these data with various parameters in the available suppliers, and set the weight for this match according to the correlation level of the matched parameters, where the weight value is related to the correlation proportional to degree level;

9) Calculate according to the weight value and the value of each parameter in the corresponding available suppliers, and sort the calculation results, and select the supplier with the highest calculation result as the selected supplier;

10) Transfer the demand information and corresponding security key to the selected supplier, record the transfer information at the same time, and transfer the transfer information, contact information and corresponding security key to the server for filing.

Preferably, in the supplier cluster analysis management and customer allocation method, the first selected generalized attribute includes a combination of the supplier's scale index, product failure rate, field bias index and supply quantity index .

Preferably, in the supplier cluster analysis management and customer allocation method, the second selected generalized attribute includes arrival time rate index, service favorable rate index, geographical area information, logistics speed information and technology R&D strength indicators.

Preferably, in the supplier cluster analysis management and customer allocation method, the predetermined lower threshold is the value obtained by taking the smallest 10% of the grade values in all supplier categories as nodes.

Preferably, in the supplier cluster analysis management and customer allocation method, the degree of relevance of a certain parameter is directly proportional to the number of other parameters associated with this parameter.

The supplier cluster analysis management and customer allocation method disclosed by the invention can automatically analyze the status of suppliers by using a computer system, and allocate customer resources, so as to realize efficient and fair management and allocation.

Detailed ways

The present invention will be described in further detail below, so that those skilled in the art can implement it with reference to the description.

The invention discloses a supplier cluster analysis management and customer allocation method. Firstly, cluster analysis is performed on suppliers, and then customer allocation is performed according to the analysis results. The purpose is to realize fairness to order customers through the cluster analysis results. distribute.

Specifically include the following steps:

1) Use the attribute-oriented induction algorithm to perform cluster analysis on all suppliers according to the first selected induction attribute, and generate the first type description of each type of supplier after clustering; the attribute-oriented induction algorithm AOI is based on the specified attribute Divide, classify all suppliers, and divide them into multiple categories.

2) Using the attribute-oriented inductive algorithm again, according to the second selected inductive attribute, perform cluster analysis on all suppliers, and generate the second type description of each type of supplier after clustering; that is to say, use two kinds of logic , to classify all suppliers, so as to avoid generalization to the greatest extent.

3) The control system sets the first evaluation rule according to the description of the first type, sets the second evaluation rule according to the description of the second type, and respectively evaluates all supplier types according to the first evaluation rule and the second evaluation rule. Carry out grade assessment; due to the different classification methods, there must be differences in the assessment rules.

4) Extract all the suppliers in the supplier category with the first category description whose level value is lower than the predetermined lower threshold, retrieve the level value of the corresponding second category description, and judge its level value Whether it is still lower than the predetermined lower threshold; the determination of the lower threshold can be determined based on the percentage, or can be determined based on the value itself.

5) Put the supplier profile data whose grade value is lower than the predetermined lower threshold described by the first category and whose grade value is still lower than the predetermined lower threshold described by the second category into the standby supplier database, and put the remaining suppliers The supplier profile data is placed in the available suppliers database; this distinguishes the available suppliers from the alternate suppliers. Alternative suppliers are assigned to customers by manual designation, while available suppliers are assigned by the system automatically.

6) Excavate the data and data of each parameter in the available suppliers, discover the correlation degree between each parameter, and evaluate the correlation degree level of each parameter; through the correlation degree evaluation, the unequal status of different parameters can be established .

7) Use the distribution processing center to receive customer information, decompose the customer information into demand information and contact information, encrypt the demand information and contact information separately, and generate their own security keys;

8) Excavate the data in the demand information, match these data with various parameters in the available suppliers, and set the weight for this match according to the correlation level of the matched parameters, where the weight value is related to the correlation The degree level is proportional; the degree of correlation is different, and the weight is different. For example, the shipment rate indicator may be more correlated with other parameters, so its weight should be higher.

9) Calculate according to the weight value and the value of each parameter in the corresponding available suppliers, and sort the calculation results, and select the supplier with the highest calculation result as the selected supplier;

10) Transfer the demand information and corresponding security key to the selected supplier, record the transfer information at the same time, and transfer the transfer information, contact information and corresponding security key to the server for filing.

Preferably, in the supplier cluster analysis management and customer allocation method, the first selected generalized attribute includes a combination of the supplier's scale index, product failure rate, field bias index and supply quantity index .

Preferably, in the supplier cluster analysis management and customer allocation method, the second selected generalized attribute includes arrival time rate index, service favorable rate index, geographical area information, logistics speed information and technology R&D strength indicators.

Preferably, in the supplier cluster analysis management and customer allocation method, the predetermined lower threshold is the value obtained by taking the smallest 10% of the grade values in all supplier categories as nodes. That is to say, rank values are queued in descending order, and the last 10% are assigned to the standby supplier sequence.

Preferably, in the supplier cluster analysis management and customer allocation method, the degree of relevance of a certain parameter is directly proportional to the number of other parameters associated with this parameter. That is to say, the more other parameters associated with this parameter, the higher the association level of this parameter.

Although the embodiment of the present invention has been disclosed as above, it is not limited to the use listed in the specification and implementation, it can be applied to various fields suitable for the present invention, and it can be easily understood by those skilled in the art Therefore, the invention is not limited to the specific details and examples shown and described herein without departing from the general concept defined by the claims and their equivalents.

Claims (5)

1. A supplier cluster analysis management and customer The user allocation method is characterized in that, firstly, cluster analysis is performed on suppliers, and then customer allocation is performed according to the analysis results, Specifically include the following steps: 1) Using the attribute-oriented induction algorithm, according to the first selected induction attribute, perform cluster analysis on all suppliers, and generate the first type description of each type of supplier after clustering; 2) Using the attribute-oriented inductive algorithm again, according to the second selected inductive attribute, perform cluster analysis on all suppliers, and generate the second type description of each type of supplier after clustering; 3) The control system sets the first evaluation rule according to the description of the first type, sets the second evaluation rule according to the description of the second type, and respectively evaluates all supplier types according to the first evaluation rule and the second evaluation rule. carry out ratings; 4) Extract all the suppliers in the supplier category with the first category description whose level value is lower than the predetermined lower threshold, retrieve the level value of the corresponding second category description, and judge its level value Whether it is still below the predetermined lower threshold; 5) Put the supplier profile data whose grade value is lower than the predetermined lower threshold described by the first category and whose grade value is still lower than the predetermined lower threshold described by the second category into the standby supplier database, and put the remaining suppliers Put the supplier profile data into the available supplier database; 6) Excavate the data and data of each parameter in the available suppliers, discover the correlation degree between each parameter, and evaluate the correlation degree level of each parameter; 7) Use the distribution processing center to receive customer information, decompose the customer information into demand information and contact information, encrypt the demand information and contact information separately, and generate their own security keys; 8) Excavate the data in the demand information, match these data with various parameters in the available suppliers, and set the weight for this match according to the correlation level of the matched parameters, where the weight value is related to the correlation proportional to degree level; 9) Calculate according to the weight value and the value of each parameter in the corresponding available suppliers, and sort the calculation results, and select the supplier with the highest calculation result as the selected supplier; 10) Transfer the demand information and corresponding security key to the selected supplier, record the transfer information at the same time, and transfer the transfer information, contact information and corresponding security key to the server for filing. 2. The supplier cluster analysis management and customer assignment method as claimed in claim 1, wherein, the first selected generalized attribute comprises the supplier's scale index, product failure rate, field bias index and supplier's index. A combination of volume indicators. 3. The supplier cluster analysis management and customer assignment method as claimed in claim 1 or 2, characterized in that, the second selected inductive attributes include arrival time rate index, service favorable rate index, geographical area Information, logistics speed information and technical research and development strength indicators. 4. The supplier cluster analysis management and customer assignment method as claimed in claim 1, wherein the predetermined lower threshold is the smallest 10% of the grade values in all supplier categories as nodes, and the obtained value. 5. The supplier cluster analysis management and customer allocation method according to claim 1, characterized in that the degree of relevance of a parameter is proportional to the number of other parameters associated with the parameter.