Supply Chain Failure Prediction Using Archived Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Supply chain planning processes often fail to synchronize due to differing data requirements and planning periods, leading to service level failures and inability to meet customer service targets, as variations in demand or production are detected too late to be accounted for in globally-optimized plans.
Innovation Solution
A system utilizing machine learning techniques on archived supply chain data to predict failures, generate alerts, and provide contextual visualizations, allowing for proactive measures without real-time data access or integration with planning and execution processes, thereby reducing computational costs and implementation time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If distinct and dissimilar processes are used for supply chain planning, then each process can operate independently with its own data requirements and planning periods, but the processes cannot be synchronized and globally-optimized plans cannot be generated
Solution Approach 1:
The patent segments the supply chain planning into distinct processes (demand planning, production planning, supply planning, distribution planning) that can operate independently, while introducing a separate prediction system that monitors failures across these segmented processes without requiring their synchronization
Solution Approach 2:
The patent introduces a machine learning-based prediction system as an intermediary layer that sits between the distinct planning processes and the service level targets. This intermediary predicts failures by analyzing historical data and patterns, enabling service level achievement without requiring the underlying processes to be synchronized or integrated
2Reliability
If real-time data access and integration with planning and execution processes are implemented, then accurate prediction and prevention of failures can be achieved, but computational costs and implementation time increase significantly
Solution Approach 1:
The patent performs preliminary action by training machine learning models on historical supply chain data in advance. The models learn patterns and relationships from past data, enabling them to predict failures without requiring real-time data access or integration with current planning and execution processes
Solution Approach 2:
The patent extracts the prediction function from the operational planning and execution systems. By separating the prediction system from the real-time processes, the patent achieves failure prediction without the computational overhead and integration complexity of real-time data access and processing
Data Source
AI summary
A system and method are disclosed for receiving only historical supply chain data from an archiving system for one or more supply chain entities storing items at stocking locations, predicting one or more supply chain events during a prediction period by applying a prediction model to a sample of historical supply chain data, calculating an occurrence risk score for at least one of the one or more supply chain events and indicating a possibility that the at least one of the one or more supply chain events will occur, generating one or more alerts identifying at least one item and at least one alert stocking location, rendering an alert heatmap visualization comprising one or more selectable user interface elements, and provide one or more tools for initiating corrective actions to be undertaken in order to resolve one or more underlying causes of the at least one alert supply chain event.