Queue Waiting Time Prediction Using Average Passage Frequency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing waiting time prediction systems for non-constant service times, such as taxi lines, face inaccuracies due to variable exit rates, leading to divergent and incorrect waiting time calculations when the number of waiting people is small.
Innovation Solution
An information processing apparatus that includes a first count unit for counting objects in a predetermined area, a second count unit for counting objects passing through a position within a unit time, and calculation units to adjust the waiting time calculation based on a threshold, using a second passage frequency when the count is below the threshold to prevent divergence and improve prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the number of waiting people is counted by analyzing captured images, then the waiting line length can be estimated, but incorrect counting occurs when the number of waiting people is small, leading to divergent waiting time predictions
Solution Approach 1:
The patent changes the calculation parameter from using real-time passing frequency to using average passage frequency calculated over a predetermined time period when the waiting line length is below the threshold. This parameter substitution stabilizes the denominator in the waiting time calculation, preventing divergence and improving prediction reliability for small queue sizes.
Solution Approach 2:
The patent dynamically adjusts the calculation method based on the waiting line length threshold. When the queue is long (above threshold), it uses real-time passing frequency for responsiveness. When the queue is short (below threshold), it switches to average passage frequency for stability. This dynamic adaptation resolves the contradiction between measurement precision and prediction reliability.
2Productivity
If the waiting time is calculated by dividing the waiting line length by the number of people leaving per unit time, then real-time waiting time can be predicted, but the prediction accuracy reduces when the number of people leaving per unit time is small
Solution Approach 1:
The patent substitutes the real-time passing frequency parameter with the average passage frequency parameter when the real-time value is too small (below threshold). This prevents the denominator from being too small, which would cause excessive sensitivity to counting errors and reduce prediction accuracy. The average value provides a more stable basis for calculation.
Solution Approach 2:
The patent pre-calculates and stores the average passage frequency over a predetermined time period as a cushion against future prediction errors. When real-time passing frequency is insufficient, this pre-computed average acts as a buffer, preventing prediction divergence and maintaining accuracy even with small queue sizes.
3Quantity of substance
If the number of people leaving the waiting line per unit time is calculated by counting passing objects, then the service rate can be determined, but incorrect predictions occur when both waiting people and leaving people counts are small
Solution Approach 1:
The patent performs preliminary action by calculating and storing the average passage frequency over a predetermined time period before using it for waiting time predictions. This pre-computation ensures that when real-time counts are too small to be reliable, a pre-established average value is available to maintain prediction accuracy, preventing the divergence that occurs with small sample sizes.
Data Source
AI summary
An information processing apparatus predicts a waiting time by suppressing a reduction in accuracy of waiting-time prediction even if the number of people waiting in a line is small. The information processing apparatus includes a calculation unit that calculates a waiting time, based on a first set of counted objects in a case where the number of first set of counted objects is less than a predetermined number and based on a second set of counted objects in a case where the first set of counted objects is greater than or equal to the predetermined number.