AI generated route data pre - processing for faster decision making

The industry is constantly generating new data, allowing take advantage of their use to increase manufacturing efficiency through the reutilisation of operational times.One such possibility is to use Artificial Intelligence (AI). While AI is heralded as a solution for industry efficiency and accuracy quandaries through process automation, its implementation grapples with the issue of inaccurate production data used for training, giving an erroneous estimation.This article introduces a framework to facilitate the integration of AI tools within companies. The focus is on the accuracy of predicting normative times and selecting alternative routes when accuracy is low. This approach improves the price quotation prediction process, where using AI is practical due to time constraints. Making quick and accurate decisions is crucial for maintaining efficiency in production.To elucidate the functionality, the authors have performed a case study in production for the framework’s validation. The wood production sector was selected due to its lower operational complexity than machinery construction, resulting in time savings in the planning process.The study revealed that the prediction accuracy of AI was 70% on average for operations normative times. To improve this accuracy, the t-test method to classify operations was used. This approach eliminates operations with low AI prediction accuracy when making price quotations. Through the case study, AI module accuracy reached 80%, while the additional data collection volume decreased by one-third.