A new research poster from the School of Continuous Improvement matches a term of dining-hall till tallies against the promised forty-dish rotation, finding three dishes absorb most of the demand and eleven are never rung.
A fourteen-week till-tally audit finds three of the roster's forty dishes drawing most of the term's covers, and eleven never rung at all
The roster was built to spread demand evenly across forty dishes. The till has never agreed to that plan.
— Dr Joost Nwosu, Postdoctoral Fellow, School of Continuous Improvement
The dining hall’s fortnightly rotation roster promises forty dishes cycling through in equal turn, each rostered for an even 2.5% share of a term’s covers. A new research poster from the School of Continuous Improvement set the roster against fourteen weeks of the servery’s own till tallies to see whether the tray agrees.
Dish-code exports were matched week by week against the published rotation, with no adjustment for roster reshuffles, of which there were none logged across the term.
Three dishes — the roast of the day, the stir-fry station, and the butter chicken — absorbed 41% of the term’s covers between them, against a combined rostered share of 7.5%. The other thirty-seven, rostered for 92.5% combined, returned 59%; eleven never reached a tray at all.
“An average that’s even across forty dishes and an average that’s even across three tell very different stories,” said Lecturer Dagny Okafor, convenor of the School’s Evaluation of Evaluation program. “The roster has been quietly averaging the first one out.”
The School counts the audit among a run of findings turning up how unevenly an even-sounding target actually lands once it reaches a term of real service.
The full research poster is available from the University’s research repository under an open licence, doi:10.5555/slop.11xj6a.