Product data analysis method and system based on maintenance work orders

Abstract

The invention provides a product data analysis method and system based on maintenance work orders, and the method comprises the steps: inputting each to-be-analyzed maintenance work order of a product into a named entity recognition model, and outputting an entity label of a fault part in each to-be-analyzed maintenance work order and an entity label of a fault phenomenon corresponding to the fault part; inputting a fault maintenance related text in each to-be-analyzed maintenance work order into a classification model, and outputting a fault reason category corresponding to the fault part; and according to the fault phenomenon extracted from the maintenance work order to be analyzed and the entity label of the fault part, in combination with the fault reason category, obtaining analysis results such as product vulnerable parts, maintenance, process optimization recommendation and the like for assisting in product and service optimization. According to the method, fault parts and fault phenomenon entity labels are automatically extracted, fault reason classification is automatically carried out, suggestions for assisting in product optimization are constructed through data mining and analysis, and a decision maker is helped to carry out product and service optimization and improvement.

Description

technical field [0001] The invention relates to the technical field of data mining, in particular to a product data analysis method and system based on maintenance work orders. Background technique [0002] In the fault maintenance and after-sales service of equipment in the industrial field, a lot of valuable data information has been accumulated. Therefore, content related to product failure and maintenance can be obtained from these data information, and then relevant personnel can be guided to maintain and maintain the product. [0003] However, most of the data such as fault phenomena and fault causes in industrial data are descriptive text data. Recorded data such as fault symptoms and fault causes vary from person to person and present unstructured features. [0004] It is not only difficult to analyze industrial data manually to extract effective information related to product failure and maintenance, but also the amount of information in text data is relatively la...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a product data analysis method and system based on maintenance work orders to solve the problem of manually extracting effective information related to product failure and maintenance from text data in the prior art, which is difficult and requires a lot of manpower Realize automatic analysis of product data to accurately obtain effective information related to product failure and maintenance

Images