CN-122019779-A - Intelligent classification system and method for claim worksheets of truck drivers

CN122019779ACN 122019779 ACN122019779 ACN 122019779ACN-122019779-A

Abstract

The invention relates to the technical field of intelligent classification, and particularly discloses an intelligent classification method for a freight car driver complaint work order, which comprises the following steps of S1, collecting work order data of a plurality of years, extracting complaint date and text, comparing and completing work order classification through multi-field keywords, S2, dividing similar work orders, setting time thresholds to distinguish urgent work orders and general work orders, counting the duty ratio of the two types of work orders, S3, calculating a priority coefficient by combining weights and the duty ratio, introducing overdue work order data correction coefficients, quantifying the work order priority, S4, periodically acquiring new work orders, classifying according to the correction coefficients, calculating a processing coefficient by combining reporting time and completing work order sorting, S5, distributing processing resources according to the duty ratio of the work order grade number, and directly processing if the work order amount does not exceed the upper limit. The method solves the problems that a truck driver demands worksheet classification, priority judgment and resource allocation pain points, realizes worksheet classification and quantitative classification through multi-field matching and coefficient correction, dynamically allocates resources and improves processing efficiency.

Inventors

WANG QIHANG
ZHOU CHENG
LIU LU

Assignees

交通运输部公路科学研究所

Dates

Publication Date: 20260512
Application Date: 20260126

Claims (8)

1. The intelligent classification method for the claim worksheets of the truck drivers is characterized by comprising the following steps of: collecting a work order of the recent year I and obtaining corresponding work order data, wherein the work order data comprises a work order complaint date r1 and a work order complaint text, and the year I represents a preset year number; keyword matching is carried out on the collected work order data according to the text of the worker Shan Suqiu, and the method comprises the following steps: And acquiring the word number N of the keywords, acquiring the i-th to i+N-th fields of the work order complaint text, recording the fields as fields to be compared, comparing the fields to be compared with the keywords, and if the fields to be compared are the same as the keywords, the work order belongs to the category of the corresponding keywords, and repeating the operation until the comparison of the whole work order complaint text is completed.
2. The method for intelligently classifying a claim worksheet of a truck driver according to claim 1, comprising: recording the worksheets classified by the same keywords as similar worksheets, presetting a time threshold value Ts, and carrying out the following operations on the similar worksheets: Calculating the processing time T=t2-T1 of similar worksheets, wherein T2 represents the date of worksheets in service, if the processing time T is more than or equal to Ts, marking the worksheets as general worksheets, and if the processing time T is less than Ts, marking the worksheets as urgent worksheets; And calculating the emergency work order occupation ratio B1=M1/M and the general work order occupation ratio B2=M2/M, wherein M1 and M2 respectively represent the number of the emergency work orders and the general work orders in the same type work orders, and M represents the total number of the same type work orders in the reported work orders.
3. The method for intelligently classifying a claim worksheet of a truck driver according to claim 2, comprising: Calculating priority coefficients Y=λ1×B1+λ2×B2+M/M all of similar worksheets, wherein M all represents the total number of reported worksheets in the recent I year, λ1 represents a preset first weight threshold value, λ2 represents a preset second weight threshold value, and λ1> λ 2>1; The method for correcting the priority coefficient Y of the similar worksheets to obtain the correction coefficient X comprises the following steps: And recording the similar worksheets with the processing time T being more than or equal to 2Ts as overdue worksheets, acquiring the number Q of the overdue worksheets, and calculating a correction coefficient X=Y-Q/M all .
4. A method for intelligent classification of claim 3, characterized by comprising the following steps: Presetting a detection period H, periodically acquiring a new report work order, comparing keywords of the new report work order, dividing similar work orders, and dividing one class of work orders and two classes of work orders based on correction coefficients corresponding to the similar work orders; For one class of work order and two classes of work order the following operations are sequentially carried out: And obtaining a correction coefficient X1 corresponding to the class of worksheets and the reported time length time1 of the class of worksheets, calculating a processing coefficient C=X1×time1/H, sequencing the class of worksheets according to the sequence of the processing coefficient C from high to low, and repeating the operations of the class of worksheets.
5. A method for intelligent classification of claim 4, characterized by comprising the following steps: And (3) obtaining the maximum work order processing quantity Max in the current detection period H, obtaining the quantity I1 and I2 of one type of work order and two types of work orders, dividing the first type of work order with larger processing coefficient C of the previous [ I1/(I1 + I2) x Max ] piece and the second type of work order with larger processing coefficient C of the previous [ I2/(I1 + I2) x Max ] piece, and carrying out priority processing.
6. The intelligent classification method for claim 4, wherein the method for classifying the class I worksheets and the class II worksheets based on the correction coefficients corresponding to the class II worksheets comprises the following steps: The similar work orders are ordered according to the order of the correction coefficient X from high to low, the first 50% of the similar work orders are marked as one type of work orders, and the rest similar work orders are marked as two types of work orders.
7. The intelligent classification method for claim 5, wherein if the number of newly reported worksheets in the detection period H is less than or equal to the maximum worksheet processing amount Max, the processing is directly performed, and the processing step is omitted.
8. An intelligent classification system for a claim work order of a truck driver, comprising: The data summarizing module is used for acquiring a work order of the recent year I and acquiring corresponding work order data, wherein the work order data comprises a work order complaint date r1 and a work order complaint text, and the year I represents the preset year number; keyword matching is carried out on the collected work order data according to the text of the worker Shan Suqiu, and the method comprises the following steps: Acquiring the word number N of the keywords, acquiring the i-th to i+N-th fields of the work order complaint text, marking the fields as fields to be compared, comparing the fields to be compared with the keywords, if the fields to be compared are the same as the keywords, the work order belongs to the category of the corresponding keywords, and repeating the operation until the comparison of the whole work order complaint text is completed; The preprocessing module is used for marking the worksheets classified by the same keywords as similar worksheets, presetting a time threshold value Ts and carrying out the following operations on the similar worksheets: Calculating the processing time T=t2-T1 of similar worksheets, wherein T2 represents the date of worksheets in service, if the processing time T is more than or equal to Ts, marking the worksheets as general worksheets, and if the processing time T is less than Ts, marking the worksheets as urgent worksheets; Calculating an emergency work order occupation ratio B1=M1/M and a general work order occupation ratio B2=M2/M, wherein M1 and M2 respectively represent the number of the emergency work orders and the general work orders in the similar work orders, and M represents the total number of the similar work orders in the reported work orders; the correction module calculates priority coefficients Y=λ1×B1+λ2×B2+M/M all of similar worksheets, wherein M all represents the total number of reported worksheets in the recent I year, λ1 represents a preset first weight threshold value, λ2 represents a preset second weight threshold value, and λ1> λ 2>1; The method for correcting the priority coefficient Y of the similar worksheets to obtain the correction coefficient X comprises the following steps: the similar work orders with the processing time T being more than or equal to 2Ts are marked as overdue work orders, the number Q of the overdue work orders is obtained, and a correction coefficient X=Y-Q/M all is calculated; The sequencing module is used for presetting a detection period H, periodically acquiring a new report work order, carrying out keyword comparison on the new report work order, dividing the same type work order, and dividing a first type work order and a second type work order based on correction coefficients of the same type work order; For one class of work order and two classes of work order the following operations are sequentially carried out: Obtaining a correction coefficient X1 corresponding to one type of work order and the reported time length time1 of the one type of work order, calculating a processing coefficient C=X1×time1/H, and sequencing the one type of work order according to the sequence of the processing coefficient C from high to low; The processing module is used for acquiring the maximum work order processing quantity Max in the current detection period H, acquiring the quantity I1 and I2 of one type of work order and two types of work orders, dividing the first type of work order with larger processing coefficient C of the previous [ I1/(I1 + I2) x Max ] piece and the second type of work order with larger processing coefficient C of the previous [ I2/(I1 + I2) x Max ] piece, and carrying out priority processing.

Description

Intelligent classification system and method for claim worksheets of truck drivers Technical Field The invention relates to the technical field of intelligent classification, in particular to an intelligent classification system and method for a freight car driver to resort to work orders. Background With the rapid development of the highway freight industry, the population scale of the freight car drivers is continuously enlarged, various requirements such as road traffic, cargo handling, vehicle maintenance, rights and interests guarantee and the like facing the transportation are increasingly increased, and the number of corresponding requirements work orders is increased in an explosive manner. The work order is used as a core carrier for receiving the demands of truck drivers and solving the problems in a coordinated manner, and the processing efficiency of the work order is directly related to the travel experience of the drivers and the service quality of the freight industry. At present, a freight management platform needs to process a large number of worksheets from different areas and different scenes, the worksheet appeal text is various and diverse, various specific problem descriptions are covered, and how to quickly and accurately classify the worksheets and process urgent important appeal preferentially becomes a key problem to be solved urgently in the industry. The existing worksheet classification method is mostly dependent on manual labeling or simple keyword matching, and lacks quantitative evaluation on the urgency degree and importance of worksheets. The traditional keyword matching can only realize basic classification, cannot judge priority by combining multi-dimensional information such as work order processing time length, similar work order occupation ratio and the like, and the situations that an emergency resort work order is delayed and a secondary work order occupies a large amount of resources often occur, so that the work order processing resource is unreasonable to be distributed, the core resort solution is not timely, and the satisfaction degree and industry management efficiency of a truck driver are affected. In order to optimize the work order processing flow and improve the classification accuracy and the resource allocation rationality, an intelligent method for classifying the work orders is required by a truck driver. Disclosure of Invention The invention aims to provide an intelligent classification system and method for a work order complaint of a truck driver, and the technical problems are solved. The aim of the invention can be achieved by the following technical scheme: An intelligent classification method for a freight car driver to resort to worksheets comprises the following steps: collecting a work order of the recent year I and obtaining corresponding work order data, wherein the work order data comprises a work order complaint date r1 and a work order complaint text, and the year I represents a preset year number; keyword matching is carried out on the collected work order data according to the text of the worker Shan Suqiu, and the method comprises the following steps: And acquiring the word number N of the keywords, acquiring the i-th to i+N-th fields of the work order complaint text, recording the fields as fields to be compared, comparing the fields to be compared with the keywords, and if the fields to be compared are the same as the keywords, the work order belongs to the category of the corresponding keywords, and repeating the operation until the comparison of the whole work order complaint text is completed. The invention further provides a scheme comprising the following steps: recording the worksheets classified by the same keywords as similar worksheets, presetting a time threshold value Ts, and carrying out the following operations on the similar worksheets: Calculating the processing time T=t2-T1 of similar worksheets, wherein T2 represents the date of worksheets in service, if the processing time T is more than or equal to Ts, marking the worksheets as general worksheets, and if the processing time T is less than Ts, marking the worksheets as urgent worksheets; And calculating the emergency work order occupation ratio B1=M1/M and the general work order occupation ratio B2=M2/M, wherein M1 and M2 respectively represent the number of the emergency work orders and the general work orders in the same type work orders, and M represents the total number of the same type work orders in the reported work orders. The invention further provides a scheme comprising the following steps: Calculating priority coefficients Y=λ1×B1+λ2×B2+M/M all of similar worksheets, wherein M all represents the total number of reported worksheets in the recent I year, λ1 represents a preset first weight threshold value, λ2 represents a preset second weight threshold value, and λ1> λ 2>1; The method for correcting the priority coefficient Y of the similar worksheets to