Wait Time Estimation Using Sensor-Based Entity Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wait time estimation techniques for customer service queues are inaccurate, especially in hybrid environments where both solo and group customers are served, as they rely solely on queue length or count, failing to provide a satisfactory representation of actual wait times.
Innovation Solution
A computer-assisted method that uses sensor data to identify single-person and multi-person waiting entities, computes estimated wait times based on their distribution and nominal wait times, and considers the number of active service points, providing an accurate estimate accessible to service-seeking individuals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If wait time estimation is based purely on queue length or count of people, then the estimation process is simple, but the accuracy of wait time estimation deteriorates in hybrid customer service environments
Solution Approach 1:
The patent segments the queue into multiple waiting entities based on sensor data, where each entity represents a group of people waiting together. This segmentation allows the system to differentiate between solo customers and group customers, who have different service time requirements. By counting entities rather than individual people, the system achieves more accurate wait time estimation without excessive complexity.
2Measurement precision
If the system accounts for the composition of waiting entities and active service points, then wait time estimation accuracy improves, but the system complexity increases
Solution Approach 1:
The system uses sensors to automatically detect and classify waiting entities in the queue without requiring manual input from customers or service staff. The sensors autonomously count entities, determine their composition (solo vs. group), and feed this data to the computation device. This self-service approach reduces operational complexity while maintaining high estimation accuracy.
Solution Approach 2:
The patent changes the fundamental parameter used for wait time estimation from individual person count to waiting entity composition. By considering the distribution of entity sizes (how many solo customers versus groups of various sizes) and the number of active service points, the system creates a more accurate model that reflects actual service dynamics without requiring complex infrastructure.
3Extent of automation
If the system uses sensor data to identify waiting entities and compute estimates, then automation increases leading to greater savings, but the difficulty of detecting and measuring queue composition increases
Solution Approach 1:
The patent replaces manual queue monitoring with sensor-based detection systems. These sensors automatically capture queue composition data, eliminating the need for human observers to physically count and classify customers. The sensors transmit this data to the computation device, which processes it to generate wait time estimates. This substitution of mechanical/sensor-based systems for manual methods achieves high automation while managing the complexity of detection through standardized sensing technology.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for estimating a wait time associated with a queue serviced by active service points. The method comprises obtaining sensor data relating to the queue; from which a plurality of waiting entities occupying respective positions in the queue is identified, each waiting entity associated with a respective number of individuals; receiving a request to compute an estimated wait time (EWT) for a given one of the individuals; computing the estimated wait time for the given individual; and causing a message comprising the EWT to be made accessible to a communication device associated with the given individual. The EWT is computed based on: (i) a distribution, in terms of the number of individuals per waiting entity, of the waiting entities occupying respective positions in the queue ahead of the given individual; (ii) nominal wait times associated with respective numbers of individuals; and (iii) the number of active service points.