Aide à la conceptualisation de la prévision des besoins au travers d'un outil didactique et intuitif en vue d'alimenter les modules de MRP et de forecasting d'un ERP.

Abstract

In the context of the ERP solutions, NSI IT Software & Services is a Microsoft’s ERP solutions provider. This company has different types of customers: some big customers who use Microsoft Dynamics AX and some smaller customers like very small enterprises (VSE) and small and medium enterprises (SME) who use Microsoft Dynamics NAV. NSI has noted that some of the customers using Microsoft Dynamics NAV did not have the sufficient maturity to understand the theme of the forecasts and realize them. Moreover, it also exists a demand from some of these customers to have a tool in Microsoft Dynamics NAV that helps them to understand and compute demand forecasts. The final objective of this project-thesis is to obtain a functional didactic and intuitive tool for the final users. Firstly, this project-thesis is focused on the analysis of the existing solutions and their features and also on the identification of the customers’ needs with the interview of two NSI’s project leaders. Secondly, the different types of forecasting methods are exposed and explained but the list of the forecasting methods in this project-thesis is non-exhaustive. The quantitative forecasting methods from the time series analysis (exponential smoothing, moving average, linear regression) are chosen to realize the functional didactic and intuitive forecasting tool. Finally, the last step of this project thesis is the development and the conceptualization of the forecasting tool. This forecasting tool has to be intuitive and didactic. Now this tool is developed in an Excel format but it can be developed later as an add-on or an extension for Microsoft Dynamics NAV to be fully-integrated. The tool appears as follows: the first step is a questionnaire for the user; the different elements and their effect are explained. Then, with the user’s answers the tool is able to attribute a forecasting method thanks to possible answers combinations.