Advancing organizational analytics : a strategic roadmap for implementing machine learning in warehouse management system

Abstract

This thesis explores how advanced analytics, specifically machine learning, can be utilized to classify spare parts, aiding the case company in optimizing space allocation and packing in warehousing. The study focuses on an organizational branch involved in logistics operations, aiming to address the issue of missing dimensional data in the company's enterprise resource planning system. By substituting this data with a classification system, the logistics stakeholders can enhance and streamline existing processes for better customer service.

A comprehensive literature review covers warehouse management systems, machine learning applications in supply chains, and analytical maturity. The research employs design science methodology, involving the creation of an innovative artifact to tackle the company's current challenges. Data collection was conducted through interviews with various employees to gather insights into their perspectives on analytical maturity, attitudes toward artificial intelligence and machine learning, and the lack of dimensional data. This qualitative data provided an understanding of the current analytical environment and resources to accommodate an advanced analytical solution.

The resulting artifact is a roadmap with a high-level machine learning model to assist with dimensional data. The findings suggest that this artifact can serve as a valuable guideline for advancing the company's analytical capabilities. It highlights key areas for supporting machine learning development and deployment. However, the study encountered two significant limitations: the abstract nature of the artifact necessitates implementation for proper evaluation, and the human resources aspect of analytical maturity was not considered in the roadmap due to insufficient data.