Online Concierge Fulfillment Time Prediction Using Segmented Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional online concierge systems fail to accurately determine if an order can be fulfilled before a retailer's closing time, leading to users being unable to select suitable time intervals, which reduces order fulfillment capacity and user satisfaction.
Innovation Solution
The system employs a fulfillment time model and a travel time model, trained through backpropagation using prior orders, to predict the time required for order completion and item delivery, and applies a buffer to account for errors, enabling or disabling time interval options based on these predictions and retailer hours of operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the system uses conventional order fulfillment time estimation methods that do not differentiate between inside retailer and outside retailer portions, then the system can maintain simple estimation logic, but the number of orders that can be fulfilled in a day is reduced
Solution Approach 1:
The patent segments the order fulfillment process into two distinct portions: (1) time spent inside the retailer obtaining items, and (2) time spent outside the retailer delivering items to the user. By separating these portions, the system can apply different estimation logic to each - using retailer operating hours for the inside portion and travel time models for the outside portion - thereby increasing the number of fulfillable orders without creating an intractably complex system.
2Adaptability or versatility
If the system prevents users from selecting time intervals where estimated arrival is later than retailer closing time, then the system ensures items are obtained before closing, but users lose the ability to select valid time intervals where pickers can obtain items before closing but deliver after closing
Solution Approach 1:
The patent separates the fulfillment time into inside-retailer time (for obtaining items) and outside-retailer time (for delivery). By this segmentation, the system allows users to select time intervals where delivery occurs after closing time, while simultaneously ensuring that item acquisition occurs before closing time. This resolves the contradiction by making both the adaptability (user selection flexibility) and reliability (closing time guarantee) true simultaneously.
Solution Approach 2:
The patent introduces an intermediary calculation step that determines the latest possible order identification time based on retailer closing time, predicted fulfillment time, and buffer time. This intermediary time threshold acts as a mediator that allows the system to both guarantee pre-closing item acquisition and permit post-closing delivery, thereby enabling users to select time intervals that would otherwise be incorrectly excluded.
3Measurement precision
If the system uses predicted fulfillment time that includes travel time from retailer to location, then the system provides comprehensive order completion time estimation, but the system incorrectly excludes valid time intervals where items are obtained before closing but delivered after closing
Solution Approach 1:
The patent segments the predicted fulfillment time into two components: time to obtain items from the retailer and time to deliver items to the user. By segmenting this measurement, the system can accurately determine whether item acquisition occurs before closing time while independently allowing delivery to occur after closing time. This segmentation resolves the measurement precision issue without sacrificing user adaptability.
Data Source
AI summary
An online concierge system receives an order from a user including items to obtain from a retailer for delivery to a location. A picker selects the order and obtains items from the retailer. The user selects a time interval during which items from the order are delivered to the location. To prevent the user from selecting a time interval for fulfillment the online concierge system prevents the user from selecting a time interval when a picker may be unable to obtain the items from the retailer before a closing time of the retailer. The online concierge system evaluates time intervals by subtracting a travel time for the picker travelling from the retailer to the location from a predicted fulfillment time for the order. This prevents the time for delivering items after being obtained from affecting whether a time interval may be selected.