ANN Item Selection for Warehouse Storage Optimization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

E-commerce companies face inefficiencies in fulfilling orders due to capacity limitations in regional distribution centers, leading to order splitting, longer fulfillment times, and increased costs, as existing methods focus solely on sales data without considering item relationships.

Innovation Solution

A system utilizing an Artificial Neural Network (ANN) to select items for storage in warehouses based on historical orders, maximizing the probability that all items in an order are stored in the same warehouse, considering item relationships and capacity constraints.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If items are selected based solely on high sales volume, then storage capacity utilization is improved, but order completeness deteriorates due to ignoring item relationships

Engineering Contradiction:

Improvestorage capacity utilizationVSAvoidorder completeness

Core Design Contradiction:

Quantity of substanceVSReliability

Solution Approach 1:

The patent combines multiple selection criteria (sales volume, item relationships, category associations) into a unified selection model. This merges the benefits of high-capacity utilization from sales-based selection with the benefits of order completeness from relationship-based selection, resolving the contradiction between these two objectives.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces an intermediary mechanism (the item selection model with neural network) that mediates between sales data and item relationships. This intermediary processes both types of information to produce optimized storage assignments that balance capacity utilization with order completeness, preventing direct conflict between the two criteria.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Quantity of substance

If warehouse capacity is limited, then storage cost is reduced, but order splitting increases leading to longer fulfillment time

Engineering Contradiction:

Improvewarehouse capacityVSAvoidfulfillment time

Core Design Contradiction:

Quantity of substanceVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-assigning items to warehouses based on historical order data and item relationships before actual orders arrive. This advance preparation ensures that frequently co-ordered items are already positioned in the same warehouse, preventing order splitting and reducing fulfillment time when capacity constraints exist.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements dynamic item assignment that adapts to changing order patterns and capacity constraints. The system continuously optimizes which items are stored in which warehouses based on current data, allowing flexible adjustment of storage assignments to minimize order splitting while respecting capacity limits.

Inventive Principle:

Principle #15Dynamics

3Ease of manufacture

If conventional sales-based item selection is used, then implementation simplicity is improved, but decision accuracy deteriorates by ignoring item relationships and understanding

Engineering Contradiction:

Improveimplementation simplicityVSAvoiddecision accuracy

Core Design Contradiction:

Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent replaces the simple mechanical rule-based selection system with an intelligent system using neural networks and machine learning. This substitution enables the system to automatically learn and capture complex item relationships and contextual understanding from historical data, significantly improving decision accuracy while maintaining ease of implementation through automated processing.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10565551B2System and method of determining item storage strategy

Publication Date: 2020.02.18 BEIJING JINGDONG SHANGKE INFORMATION TECH CO LTD

AI summary

A system for selecting items to be stored in a warehouse from a list of candidate items. The system includes a computing device having a processor and a storage device storing computer executable code. The executed code is configured to: assign scores to the respective candidate items based on an item selection model; and select items from the candidate items to be stored in the warehouse according to the assigned scores under a capacity limit of the warehouse. The item selection model is configured with an Artificial Neural Network (ANN), having item data for an item as input and a score for that item as output. The score represents a probability that the item is to be stored in the warehouse. The ANN is trained by historical orders with an objective of maximizing a probability that the orders have their respective item(s) all stored in the same warehouse.