Question

A suburb has 2000 households. The council hired two transport consulting companies to predict the generated work trips from within the suburb after building a few residential complexes. The new buildings will increase the number of households to 2500. The following information is provided: Type Household Size Household Count Projected Household Count 800 900 800 700 Each company has conducted different travel surveys from 200 households based on different sampling methods. The work trip generation models are as follows: Company A: Ti -0.10.8 (household size) Company B: T 0.10.5 (household size) If the total number of current work trips is 2700, which model should be used for prediction? Why? What are the possible reasons for the difference between the two models? As a transport engineer working for council, propose a new and more accurate model based on Companies A and B models. Based on the new model in Part C, predict the number of work trips after the completion of construction a. b. c. d.

Answer 1

(a) Total household surveyed: 200

given the household count the proprotion surveyed would be 50, 80 and 70

Total House hold Size =

(50 x 1) + (80 × 2) + (70 x 3) 420

Trip size from company 'A' according to regression model:

(0.8 × 420)-0.1 335.9

Trip size from company 'B' according to regression model:

(0.5 420) 0.1 210.1

Scaling to 2500 household

prediction from company B is close to 2700. Hence we should use prediction model of company B.