Proactive Request System Using Triple Moving Average Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data prediction technologies fail to accurately and reliably predict future events, especially for locations with intermittent events and recent trends, leading to inefficient resource allocation and waste in proactive response systems.
Innovation Solution
A proactive request communication system that uses a triple moving average method combining location-specific and zone-specific components, along with an improved rounding process to transform non-integer prediction values into interpretable integers, enhancing the accuracy and reliability of data prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If previous data prediction technology bases predictions on events from the same time period in the previous year, then it maintains a simple prediction model, but it fails to capture recent trends and changes
Solution Approach 1:
The patent segments the prediction model into three distinct moving average calculations: (1) location-specific moving average capturing recent trends at each location, (2) zone-specific moving average capturing regional patterns, and (3) item-type specific moving average capturing product-level trends. This segmentation allows the system to capture multiple dimensions of recent trends while maintaining manageable complexity through modular calculation components.
Solution Approach 2:
The patent adds temporal dimensionality by incorporating multiple time-weighted moving averages with different scopes (location, zone, item-type) rather than relying on a single historical period comparison. This multi-dimensional approach captures recent trends across different granularities, transforming the simple year-ago comparison into a sophisticated multi-perspective temporal analysis.
2Adaptability or versatility
If the system predicts events at locations with intermittent activity (zero or one event per day), then it can provide coverage for more locations, but the prediction reliability decreases
Solution Approach 1:
The patent introduces zone-specific moving averages as an intermediary layer between location-specific data and overall predictions. For locations with intermittent events, the zone-level aggregation provides a stabilizing influence, allowing the system to leverage regional patterns to predict individual location behavior. This intermediary approach enables coverage of sparse locations while maintaining reliability through contextual information from broader zones.
Solution Approach 2:
The patent merges three different moving average calculations (location-specific, zone-specific, and item-type specific) into a unified prediction framework. This combination allows the system to leverage multiple data sources and patterns, where the strength of one component can compensate for the weakness of another, particularly for intermittent events where location-specific data is sparse but zone-level or item-type patterns may be more reliable.
3Measurement precision
If the system uses non-integer prediction values, then it maintains prediction precision, but the values become difficult to interpret and implement
Solution Approach 1:
The patent applies rounding as a preliminary action to transform non-integer prediction values into integers before they are used for decision-making. By pre-processing the prediction output through rounding, the system maintains the precision benefits of continuous calculations while ensuring the final values are easily interpretable and actionable for inventory management and resource allocation decisions.
Data Source
AI summary
A data prediction subsystem stores event data indicating amounts of items removed from locations over a previous period of time and event-to-status transition rules for each location. An event is detected at a first location. The detected event is associated with a change in status of a first item. Based on the detected event and the event-to-status transition rules, an anticipated item status is determined for the first item, indicating whether the first item is believed to be present at the first location at a time during the previous period of time of the event data. Based at least in part on the anticipated item status for the first item, a prediction value is determined that corresponds to a recommended amount of the first item to request for a future time.