Data Value Management – Prediction and Reduction of Micro-Stops
Data Value Management
Sector: Industry
Business Case
The challenge is to develop a system capable of detecting the primary causes that provoke a large number of micro-stops in the different production lines of a company, in order to reduce their occurrence. Additionally, a secondary system is actively being sought to predict these issues in real time, providing early warnings and effectively reducing their duration.
Objectives
The objective of this challenge is to identify the primary causes that lead to the generation of a greater number of micro-stops in order to prevent them from occurring, as well as to predict those that cannot be avoided, warn the operator and reduce their duration.
Use case
The proposed solution involves a data structuring process, followed by the training of models with high explanatory power to identify the root causes of micro-stops. Additionally, models with high predictive power are developed to make accurate predictions.
Infrastructure
On Premise
Technology
Machine learning and deep learning
Data
The data used are the company’s own data including production parameters, the characteristics of the products produced and the micro-stops that occurred.
Resources
The resources used were the Data Value Management staff themselves who were in charge of preparing the data, implementing the predictive systems and training the users for the developed system.
Difficulties and learning
The main difficulties encountered in the project stemmed from the quality of the data, an area in which a large amount of resources were invested.
KPIs (business impact and metrics of the model)
The main KPIs include the percentage of micro-stops explained by the explanatory algorithms, which is 37%, and the percentage of micro-stops that can be anticipated by the predictive models, which is 60%.
Funding
No funding from any public or private entity.
Collaborators, Partners
The project was carried out exclusively by DVM, with the help of the company’s staff who provided business knowledge.
