Artificial Intelligence for Truck Fleet Management**

今天的运输服务比以往任何时候都更快，公司面临着挑战，即在维护道路安全的同时，更快地分销产品，并以越来越有竞争力的成本。随着卡车上数据收集系统的出现，为数据智能创造了巨大的机会，这与为各种事件创建解释性和预测性模型有关。该项目的目标是(1)开发数据驱动的模型，以预测卡车车队中发生的道路事故和违规行为，以及(2)确定潜在的应对措施，以减少有害事件的发生。为了建立预测模型，我们依赖于随机森林，这是一种最先进的分类和回归技术，已成功应用于许多问题，包括在供应链建模中。为了解释结果并确定对策，我们将使用数据挖掘技术，例如关联规则。我们希望这个项目能为罗伯特集团提供预测模型和可能的行动，以改善其员工在道路上的**安全，这将提高其在加拿大的竞争力。

Today's transportation services are moving faster than ever, and companies are challenged to distribute their products more quickly and with increasingly competitive costs while maintaining safety on the road. With the emergence of data collection systems in trucks, tremendous opportunities for data intelligence are created, related to the creation of explanatory and predictive models for various events. The goal of this project is (1) to develop data-driven models to predict road accidents and infractions occurring in the truck fleet, and (2) to identify potential counter-measures to reduce the incidence of detrimental events. To build the predictive models, we pan to rely on Random Forests, a state- of-the-art classification and regression technique that has been successfully applied to many problems including in supply-chain modeling. To interpret the results and identify counter measures, we will use data mining techniques, for instance Association Rules. We expect this project to provide Groupe Robert with predictive models and possible actions to improve the **safety of its employees on the road, which will improve its competitiveness inside Canada.