Intelligent Inventory Distribution System for Fulfillment Centers

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional inventory management systems for fulfillment centers (FCs) fail to adequately distribute products among multiple FCs, leading to backlog issues due to insufficient consideration of individual FC capabilities, such as inbound and outbound capacity, resulting in potential sales losses.

Innovation Solution

A computer-implemented system that determines target inventory for stock keeping units (SKUs) by calculating cycle and safety stock, generates profiles for each FC based on historical data, and assigns regional target inventory portions to FCs, ensuring optimal distribution based on their capacities and current inventory levels.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional inventory management systems distribute products to multiple fulfillment centers, then product availability may be improved, but individual fulfillment center capabilities (inbound/outbound capacity) are not adequately considered, leading to backlogs and processing delays

Engineering Contradiction:
Improveproduct availabilityVSAvoidprocessing speed
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system generates individual profiles for each fulfillment center that capture local characteristics such as inbound capacity, outbound capacity, and current inventory levels. This local quality approach ensures that product distribution decisions are tailored to the specific capabilities of each fulfillment center rather than applying a uniform distribution strategy, thereby preventing backlogs at centers with limited capacity while maintaining high product availability across the network.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system dynamically adjusts product distribution by continuously monitoring fulfillment center profiles and modifying allocation decisions based on current capacity conditions. The dynamic nature of the system allows it to adapt to changing inbound/outbound capacities and inventory levels in real-time, ensuring that distribution strategies optimize both product availability and processing speed under varying operational conditions.

Inventive Principle:
Principle #15Dynamics

2Quantity of substance

If large volumes of products are ordered for peak season, then demand coverage may be improved, but fulfillment centers may not have sufficient resources to receive products in a timely manner, creating backlogs

Engineering Contradiction:
Improveinventory volumeVSAvoidreceiving time
Core Design Contradiction:
Quantity of substanceVSLoss of time

Solution Approach 1:

The system performs preliminary assessment of fulfillment center capacities before peak season inventory allocation. By generating fulfillment center profiles that include inbound capacity metrics before the peak season arrives, the system can pre-plan receiving schedules and allocate inventory volumes that match anticipated capacity availability, preventing backlogs while ensuring sufficient inventory volume is distributed to meet peak demand.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system incorporates feedback loops that monitor actual receiving performance against predicted capacity during peak season. This feedback mechanism allows the system to adjust subsequent allocation decisions in real-time, ensuring that inventory volume increases do not exceed the actual receiving capacity of fulfillment centers, thereby preventing time losses and backlogs even as inventory volumes scale up for peak demand.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS11321651B2Cycle and safety stock determination systems
Publication Date: 2022.05.03 COUPANG CORP
  • US11321651B2 patent drawing
  • US11321651B2 patent drawing
  • US11321651B2 patent drawing

AI summary

Disclosed are methods and systems for intelligent distribution of products. In one aspect, a system comprises a memory storing instructions and at least one processor configured to execute the instructions. The processor performs operations include receiving forecasting data comprising an expected demand of a stock keeping unit and determining a target inventory of the SKU to satisfy the expected demand. The operations further include determining a regional target inventory of the SKU of a region comprising a plurality of fulfillment centers. The operations further comprise receiving historical data comprising the inbound and outbound shipment history of a fulfillment center in the plurality of fulfillment centers and generating a profile for the fulfillment center. The operations further comprise assigning a portion of the regional target inventory to the fulfillment center and transmitting instructions to a device to stock a number of SKUs in the fulfillment center based on the portion.