Dynamic Put-Away Using Digital Twins for Warehouse Congestion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current simulation and warehouse management software inefficiently assigns storage locations, fails to account for nesting items, and does not accurately predict congestion, leading to suboptimal storage density and excess handling of items.
Innovation Solution
A digital twin simulation model dynamically determines storage locations for items based on real-time factors, including nesting and congestion, using machine learning to optimize storage density and reduce handling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If storage locations are assigned based on closest placement to reduce travel time, then item retrieval speed is improved, but storage density decreases and excess handling occurs
Solution Approach 1:
The system performs preliminary actions by predicting future picking requests and pre-positioning items in optimal storage locations before they are actually needed. The simulation model forecasts demand patterns and proactively arranges items to minimize future travel time and maximize storage density, rather than reactively responding to each picking request as it occurs.
Solution Approach 2:
The system implements dynamic storage location assignment that continuously adapts to changing conditions. The simulation model dynamically adjusts put-away strategies based on real-time factors such as current storage density, predicted demand patterns, item nesting opportunities, and congestion levels, transforming static warehouse management into a dynamic optimization process.
2Quantity of substance
If simulation model accounts for nesting and congestion factors, then storage density improves, but computing resource consumption increases
Solution Approach 1:
The system applies partial action by selectively incorporating nesting and congestion factors into the simulation model only when they provide meaningful optimization value. Rather than exhaustively modeling every possible factor, the system identifies and processes the most impactful parameters (nesting coefficients, congestion zones) to achieve sufficient storage density improvement without excessive computing overhead.
Solution Approach 2:
The system utilizes parameter changes by dynamically adjusting simulation parameters such as nesting coefficients and congestion weights based on current warehouse conditions. The simulation model modifies these parameters in real-time to reflect changing storage patterns and traffic flows, enabling adaptive optimization of storage density while managing computational complexity through parameter rather than structural changes.
3Productivity
If dynamic put-away optimization is implemented, then item handling is reduced, but system complexity increases
Solution Approach 1:
The system employs copying by creating a virtual simulation model (digital twin) of the physical warehouse that mirrors its structure, operations, and constraints. This virtual copy allows the system to test and optimize put-away strategies in silico before implementing them in the physical warehouse, reducing the complexity burden on the actual warehouse management system while achieving advanced optimization capabilities.
Solution Approach 2:
The simulation model acts as an intermediary layer between the complex warehouse environment and the control system. It absorbs and processes the complexity of modeling nesting, congestion, and dynamic conditions, then presents simplified, actionable put-away instructions to the warehouse management system, effectively mediating between complexity and usability.
Data Source
AI summary
A facility for determining storage locations for items in a complex during put-away operations is described. The facility receives an indication of one or more orders that indicate a time at which at least one item is scheduled to arrive at the complex. The facility obtains simulation data that includes one or more predicted states of the complex by applying the indicated orders to a digital twin simulation of a complex. The facility identifies attributes of an item indicated in at least one of the orders and identifies one or more storage locations for the item based on the simulation data and the attributes.