Dynamic Picker Radius Expansion for Order Attractiveness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Online concierge systems face challenges in efficiently matching orders with pickers due to the fixed maximum driving distance filter, leading to inefficiencies and potential late deliveries, as more attractive orders may not require significant expansion of this filter, while less attractive orders need broader reach, and existing methods lack technical solutions to balance these factors.
Innovation Solution
An online concierge system uses a trained computer model to predict an attractiveness metric for orders, classifying them as 'attractive' or 'normal,' and adjusts the maximum driving distance filter's expansion rate based on this classification, ensuring that attractive orders are made available to a smaller, gradually expanding group of pickers and normal orders to a larger, more rapidly expanding group, thereby optimizing picker selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the maximum driving distance filter radius is expanded quickly to ensure orders are accepted in time, then the risk of late deliveries is reduced, but the matching efficiency deteriorates as farther pickers may accept orders before closer pickers
Solution Approach 1:
The system dynamically changes the maximum driving distance parameter based on order attractiveness predictions. For highly attractive orders, the radius is expanded more aggressively to ensure quick acceptance, while for less attractive orders, the expansion is more conservative to maintain matching efficiency. This parameter adaptation resolves the contradiction by adjusting the distance threshold according to the specific order's characteristics.
Solution Approach 2:
The maximum driving distance filter transitions from a static fixed threshold to a dynamic, order-specific radius that expands over time based on prediction models. The system continuously adjusts the search radius for each order individually, allowing the parameter to be flexible and adaptive rather than rigid and uniform across all orders.
2Productivity
If the maximum driving distance filter radius is kept small to maintain picker matching efficiency, then closer pickers are prioritized, but the risk of late deliveries increases for less attractive orders
Solution Approach 1:
The system adjusts the maximum driving distance parameter dynamically based on order attractiveness predictions. For orders predicted to be less attractive, the radius expansion is controlled and more conservative, maintaining efficiency while still allowing sufficient expansion to ensure timely acceptance. This resolves the contradiction by adapting the distance threshold to each order's specific characteristics.
Solution Approach 2:
The filter radius becomes a dynamic parameter that evolves over time based on real-time predictions and order acceptance patterns. Rather than maintaining a fixed small radius, the system allows the radius to expand adaptively for each order, balancing the need for efficiency with the need to ensure timely delivery acceptance.
3Ease of operation
If a fixed maximum driving distance filter is applied to all orders, then the system is simple to operate, but it cannot adapt to different order attractiveness levels
Solution Approach 1:
The system automatically adjusts the maximum driving distance parameter based on predicted order attractiveness, eliminating the need for manual configuration. The parameter changes are driven by machine learning predictions, allowing the system to adapt to different order types automatically while maintaining ease of operation through automation rather than manual intervention.
Solution Approach 2:
The system performs self-optimization by using prediction models to automatically determine appropriate radius expansions for each order without human intervention. The algorithm autonomously adjusts parameters based on order characteristics and historical data, making the system adaptive while keeping the user interface simple and requiring no manual tuning.
Data Source
AI summary
Embodiments relate to order specific expansion of an area that encompasses pickers available for accepting an order placed with an online system. The online system accesses a computer model trained to predict an attractiveness metric for the order and applies the computer model to predict a value of the attractiveness metric for a first order. The online system classifies the first order into a first set or a second set, based on the value of the attractiveness metric and a threshold. Based on the classification, the online system expands over time a size of an area that encompasses a set of pickers available for accepting the first order. The online system causes a device of each picker in the set of available pickers located within the area of the expanded size to display an availability of the first order for acceptance by each picker in the set of available pickers.