Order-Level Delivery Option Availability Using Acceptance Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional online concierge systems struggle to make individual decisions on delivery service options due to inadequate mechanisms for treating different shopping carts differently, leading to suboptimal availability of ETA delivery service options and lost customer demand during supply constraints.
Innovation Solution
A machine-learning computer model is trained to predict a metric (e.g., time to accept) for each order, enabling per-order decisions on service options, such as ETA availability, to improve fulfillment efficiency and customer experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ETA delivery service options are offered to all orders during supply constraints, then customer conversion is improved, but fulfillment efficiency deteriorates due to picker idle time
Solution Approach 1:
The system applies different service option availability decisions to different orders based on their individual characteristics (attractiveness metrics). Instead of a uniform zone-level approach, each order is evaluated locally to determine whether ETA options should be offered, allowing the system to optimize for both customer conversion and fulfillment efficiency at the individual order level.
Solution Approach 2:
The system dynamically changes the availability parameter of ETA delivery service options based on the predicted attractiveness metric of each order. By adjusting this parameter (availability) according to order-specific conditions, the system resolves the contradiction between maintaining high conversion rates and preserving fulfillment efficiency during supply constraints.
2Device complexity
If ETA delivery service option availability is determined at zone level, then system complexity is reduced, but manufacturing precision deteriorates due to inability to treat different orders individually
Solution Approach 1:
The system segments the decision-making process from the zone level down to the individual order level. By evaluating each order separately based on its attractiveness metric, the system achieves precise order-level decisions without requiring complete redesign of the zone-level architecture, thus balancing complexity and precision.
Solution Approach 2:
The system introduces dynamic, order-specific availability decisions within the existing zone-level framework. The ETA service option availability becomes a dynamic parameter that adjusts based on each order's characteristics, allowing the system to maintain simplicity at the macro level while achieving precision at the micro level.
3Productivity
If picker engagement is prioritized to reduce idle time, then fulfillment throughput is improved, but service quality deteriorates due to selective ETA availability
Solution Approach 1:
The system applies partial action by offering ETA delivery options to only some orders (those with high attractiveness metrics) rather than all orders. This selective approach ensures that limited picker resources are allocated to orders that will generate the most value, improving throughput while maintaining service quality for prioritized orders.
Solution Approach 2:
The system uses feedback from the computer model's attractiveness predictions to dynamically adjust ETA availability decisions. This feedback mechanism ensures that service quality is maintained for high-value orders while optimizing overall fulfillment throughput by being selective about which orders receive priority service options.
Data Source
AI summary
Embodiments relate to determining an availability of a service option for delivery of an order placed with an online system. The online system receives an order placed with the online system. The online system accesses a computer model trained to predict a value of metric for an order placed with the online system. The online system applies the computer model to predict the value of the metric for the order. The online system determines which service option of a plurality of service options of the online system is available for delivery of the order, based at least in part on the predicted value of the metric and a threshold. The online system causes the device of the user to display an availability of the determined service option for delivery of the order.