Loading…
Application of artificial intelligence through machine learning algorithms to reduce the number of OS in the field, based on failures found on the refrigerator assembly line in a multinational consumer goods industry/ Aplicacao de inteligencia artificial atraves de algoritmos de aprendizagem de maquina para reducao do numero de OS em campo, baseado em falhas encontradas na linha de montagem de refrigeradores em uma industria multinacional de bens de consumo
This work consists of applying artificial intelligence through machine learning algorithms to reduce work orders in a multinational consumer goods industry. The data used in this work comes from refrigerator assembly line failures and work orders generated for after-sales refrigerators within a 6-mo...
Saved in:
Published in: | GeSec : Revista de Gestão e Secretariado 2023-10, Vol.14 (10), p.18396 |
---|---|
Main Authors: | , , , |
Format: | Article |
Language: | Portuguese |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | This work consists of applying artificial intelligence through machine learning algorithms to reduce work orders in a multinational consumer goods industry. The data used in this work comes from refrigerator assembly line failures and work orders generated for after-sales refrigerators within a 6-month interval after they were manufactured. Data on refrigerator assembly failures are obtained through the industrial assembly line, which are consecutively recorded in specific databases. After that, the data is migrated to a new database and normalized for the application of AI. As well as work order data that is generated, recorded in specific databases, normalized and migrated to a new database, ready to be manipulated with AI. The separation, normalization and organization of data is carried out using the SQL language and a SQL Server database. Artificial Intelligence is used through the application of machine learning algorithms such as Random Forest, Light GBM and Linear Regression, through the Python programming language and techniques such as Grid Search and Cross Validation. Therefore, the main objective of combining these technologies is to present the main failures that contribute to the generation of work orders and the prediction of the quantity of OS that can be generated, based on the failures that occurred on the industrial refrigerator assembly line. Keywords: Assembly Line Failures. Order of Service. Refrigerators. Machine Learning. Artificial intelligence. Este trabalho consiste na aplicacao de inteligencia artificial atraves de algoritmos de aprendizagem de maquina para reducao de ordens de servico de uma industria multinacional de bens de consumo. Os dados utilizados neste trabalho sao provenientes das falhas da linha de montagem de refrigeradores e ordens de servico geradas para refrigeradores pos-venda em um intervalo de 6 meses apos serem fabricados. Os dados das falhas de montagem de refrigeradores sao obtidos atraves da linha de montagem industrial, que consecutivamente sao gravados em bases de dados especificas. Apos isso, os dados sao migrados para uma nova base de dados e normalizados para a aplicacao da IA. Assim como os dados de ordens de servico que sao gerados, gravados em bases de dados especificas, normalizados e migrados para uma nova base de dados, prontos para serem manipulados com IA. A separacao, normalizacao e organizacao dos dados e realizada atraves da linguagem SQL e uma base de dados SQL Server. Ja, a Inteligencia Ar |
---|---|
ISSN: | 2178-9010 2178-9010 |
DOI: | 10.7769/gesec.v14i10.3055 |