Errand Sequence Optimization via Visit Duration Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for planning sequences of errands are inefficient as they rely on personal experience and limited information, failing to accurately account for visit durations and travel times, leading to suboptimal trip planning and wasted time.
Innovation Solution
A computer-implemented method and system that uses consumer location data, transaction data, and travel data to predict visit durations and optimize the sequence of visits to merchant locations, providing an optimal order and route that minimizes total duration regardless of travel distance, leveraging regression algorithms and real-time data processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If consumers plan errand sequences based on personal experience and limited online information, then the planning process is simple and requires minimal data processing, but the accuracy of visit duration and travel time estimates deteriorates, leading to suboptimal trip planning
Solution Approach 1:
The patent combines multiple data sources (transaction data, consumer location data, travel data) and merges them with regression algorithms to create a comprehensive planning system. This integration of diverse data types enables accurate duration and time estimates while resolving the contradiction between measurement precision and system complexity.
Solution Approach 2:
The system performs preliminary data collection and model generation before the actual trip planning. By pre-processing transaction data, location data, and travel data to create regression models, the system prepares accurate estimation capabilities in advance, improving measurement precision without adding complexity during the actual planning moment.
2Loss of time
If consumers plan errand sequences based on travel distance alone, then the routing is simple to calculate, but the total trip duration deteriorates because visit durations are not accounted for
Solution Approach 1:
The patent changes the optimization parameter from单纯的 travel distance to total trip duration by incorporating visit duration estimates. The regression models provide parameter estimates for visit durations, which are then integrated into the sequence optimization calculation, reducing time loss while managing algorithmic complexity through data-driven parameter substitution.
Solution Approach 2:
The system uses feedback from historical transaction data and location data to continuously improve duration estimates. By analyzing past consumer behavior patterns and feeding this information back into the regression models, the system optimizes trip duration predictions without requiring exponentially more complex algorithms.
3Measurement precision
If detailed transaction data and consumer location data are collected from multiple consumers, then the accuracy of duration predictions improves, but the amount of data processing and storage requirements increases
Solution Approach 1:
The patent extracts only the essential features from large volumes of transaction and location data - specifically visit durations and patterns - rather than processing all raw data. By extracting key predictive features and feeding them into regression models, the system achieves accurate predictions while reducing the effective data volume that requires continuous processing and storage.
Data Source
AI summary
A system and method is disclosed for determining an optimized sequence of visits by a consumer to a plurality of merchant locations irrespective of travel distance. The system is configured to receive, from the consumer, errand data including start time and start and finish location and a plurality of merchant locations to visit. The system also generates one or more models for predicting visit durations the merchant locations as a function of time using transaction data and location data associated with a plurality of consumers that previously visited the merchant locations. Based on the generated models, and current and historical travel/traffic information and the errand data, the system generates an optimized sequence of visits for the consumer.