Price, Weights and Rankings: A Fuzzy AHP–TOPSIS and Data-Mining Evaluation of 186 Passenger Cars in the Iranian Market

Choosing a car is a multi-criteria decision in which technical performance competes with price. In a protected market such as Iran's in 2011, where import tariffs and two dominant domestic producers kept prices far from international levels, the weight given to price can change a ranking completely. This paper evaluates 186 passenger-car models sold in Iran against nine criteria: top speed, price, combined fuel consumption, number of valves, power-to-weight ratio, specific output, engine size, 0–100 km/h acceleration time and engine power. Criterion weights were obtained by fuzzy AHP with Chang's extent-analysis method from pairwise comparisons by 185 engineering managers of Iranian car manufacturers. Price received the highest weight (0.154), followed by engine power (0.151) and top speed (0.136). The models were ranked by TOPSIS under four scenarios that switch price and expert weights on and off. The data were then mined with k-means clustering and a C&RT decision tree. Price is the dominant driver: removing it changes the ranking far more than removing the expert weights (Spearman ρ = 0.23 between the two weighted scenarios, against ρ = 0.997 between the two price-free scenarios). Most domestic models rank mid-table only when price is included and fall to the bottom third on technical criteria alone. Without price, a decision tree reproduces the three TOPSIS score classes with a ten-fold cross-validated accuracy of 0.90, against 0.42 for a majority-class baseline. The paper is a revised English version of the author's MBA thesis (2011). All TOPSIS rankings are recomputed from the thesis data matrix; the clustering results and decision rules are those reported in the thesis.