A study in Malang measured accuracy and computation time for AHP, Fuzzy AHP, and AHP TOPSIS on the same recommendation task.
Picking where to eat sounds trivial until you have to automate it. A recommendation system has to weigh distance, price, and rating at the same time, then produce a ranked list fast enough to be useful on a phone screen.
A study from Brawijaya University compared three multi criteria methods on exactly that task, using 30 real cases from Malang, Indonesia. The methods were plain AHP, Fuzzy AHP, and AHP combined with TOPSIS.
AHP breaks a decision into a hierarchy of criteria and alternatives, then derives weights from pairwise comparisons. It is simple and widely used, but its accuracy depends on how consistently those comparisons are filled in.
Fuzzy AHP keeps the same structure but replaces crisp numbers with triangular fuzzy values. Judgments like fairly important or strongly important become ranges rather than single points, which softens the effect of subjective scoring.
AHP TOPSIS splits the job. AHP sets the criteria weights, then TOPSIS ranks the alternatives by their distance from an ideal solution. Its selling point is speed rather than accuracy.
On the 30 test cases, plain AHP reached 30 percent accuracy. AHP TOPSIS improved that to 40 percent. Fuzzy AHP was clearly ahead at 66.67 percent, more than doubling the baseline.
The authors attribute the gain to the weighting step. Fuzzy values handle vagueness in human judgment better than fixed numbers, so the criteria weights land closer to what users actually mean.
On a phone, computation cost is a first class concern. AHP and Fuzzy AHP both grow on the order of mn squared, where m is the number of criteria and n the number of alternatives. AHP TOPSIS grows quadratically in n, and it came out fastest in a simulation with three criteria and 50 to 150 menu options.
The authors conclude that all three methods are workable for a mobile culinary app. The choice depends on what you are short of. If accuracy matters most, Fuzzy AHP is the pick. If the device is slow or the list is long, AHP TOPSIS is the safer bet.
Dewi et al. (2018). A Comparison Between AHP and Hybrid AHP for Mobile Based Culinary Recommendation System. International Journal of Interactive Mobile Technologies.
