Problem Statement
Facing increasing complexity, the order picking process remains a critical, labor-intensive bottleneck in modern warehouse operations, though organizations recognize the potential of digital twin technology, virtual replicas using real-time data, to drive improvements. A fundamental challenge exists in translating this theoretical concept into a practical, demonstrable solution that specifically targets and quantifies gains in order picking performance. This research directly addresses this gap by developing a functional digital twin prototype, aiming to move beyond theoretical benefits to deliver a promising solution that uses real-time data to support decision-making, directly quantify performance improvements, and thereby enhance the crucial KPIs of efficiency, productivity, and scalability within the most demanding area of the warehouse.
Research Question
How can a digital twin prototype be developed and implemented to effectively enhance key performance indicators (KPIs) within the complex order picking process of a warehouse?
Research Approach (Methodology)
To develop the digital twin prototype, the researchers employed several methods:
- Explored Technologies: Examined the feasibility of using technologies like sensors, automated guided vehicles (AGVs), picking robots, automated storage and retrieval systems (AS/RS), and AI to improve efficiency and productivity.
- Data Collection: Conducted stakeholder interviews, process mapping, and gathered data on historical order demand and daily labor hours.
- Model Formulation: Formulated a workforce AI forecasting model harnessing machine learning techniques to predict labor requirements.
Key Results
This project culminates in a comprehensive roadmap for implementing these solutions. Leveraging the forecasting model alongside the recommended technologies will allow the warehouse team to enhance their KPIs for efficiency and productivity. The digital twin prototype has the potential for scaling to other processes and warehouses.
References