Canteen Crowd Analysis
Opens the original study.
- Study
- Queue-management UX study, survey based
- Where
- Five canteens at Dhirubhai Ambani University
- Instrument
- 11-question survey
- Method
- Nine hypotheses, two regression models, clustering
- Led to
- Canteen Management System for DAU, in progress
01 Question
Everyone knows the canteens are crowded. What does the crowd cost?
Complaints cannot separate a throughput problem from a queue-design problem or a perception problem, and each needs a different fix. 204 students answered a survey on waiting behaviour and waiting experience.
- Respondents204
- Canteens5
- Questions11
02 Method
Nine hypotheses were fixed in advance, then tested with regression and clustering.
The survey also measured appetite for two interventions: a pre-order system and a live wait-time display. Analysis applied Bonferroni correction to the nine hypotheses, then ran two regressions, a correspondence analysis and a clustering step.
- Hypotheses9
- Regression models2
- Respondents204
03 Findings
Most students wait ten minutes or more.
64% wait 10 minutes or more on a typical busy day, and 81% visit daily. The queue defines the lunch hour.
- Wait 10 minutes or more64%
- Visit daily81%
03 Findings
The wait hurts more than its length suggests.
60% have skipped a meal because of the wait. Once the experience rating is controlled for, wait duration no longer predicts skipping, so the way the wait feels drives the harm.
- Skipped a meal123
- Of respondents60%
03 Findings
Pre-ordering and live wait times appeal to the same students.
Students who want one want the other, which points to a shared need for information and control. 78% would pre-order, including 85% of those who have skipped a meal.
- Would pre-order78%
- Of meal-skippers would85%
03 Findings
Waiting pain does not translate into willingness to pay.
Only 13% would pay a convenience fee. Students with the longest waits unanimously reject one.
- Would pay a fee13%
- Of 204 students27
03 Findings
Three groups queue in three different ways.
Clustering identifies the Convenience Investors, the only group open to a fee, the Frustrated Endurers, who carry the heaviest queue burden, and the Light-Wait Pragmatists.
- Convenience Investors75
- Frustrated Endurers77
- Light-Wait Pragmatists52
Respondents
204 people
Each dot is one respondent. Dots show group sizes, not individual answers.
04 Directions
- 75 studentsThe Convenience InvestorsThe only group open to a fee, with a mean rating of 2.93 out of 5. Ready to adopt.
- 77 studentsThe Frustrated EndurersThe heaviest queue burden, a mean rating of 2.40 and unanimous opposition to a fee. 66% have skipped a meal.
- 52 studentsThe Light-Wait PragmatistsThe shortest waits, and mostly opposed to a fee.
- 01Design for the perceived waitShow order position, live queue length and a countdown to reduce uncertainty on arrival.
- 02One system, not two featuresLive load, the decision and the pre-order sit in one flow.
- 03A free core and an optional priorityThe live wait-time display stays free. Only optional add-ons carry a price.
- 04Meal-skippers are the beachheadThe pilot recruits from them and measures success as less skipping.
- 05Fix the compound and lead with the queueMarked lanes and digital tokens take bodies out of the physical crush.
- 06Design the wait itselfSeating near the queue and order-status screens serve the idle quarter.
05 Limits
- Wait times are recalled estimates, and no stopwatch measurements back them.
- The associations are correlational and do not establish cause.
- One respondent who answered Rarely was merged into the 1 to 3 times bucket, a stated analytical decision.
The research site
Get in Touch