SELECTING PATH FLOW PATTERN FOR SENSITIVITY ANALYSIS OF CAPACITATED NETWORK OF USER EQUILIBRIUM

Sensitivity analysis is a means for extracting the cause and effect relationship between inputs and outputs in the network model. The paper proposes the scheme of integrating column generation and penalization of capacity constraints, to find a set of equilibrium path flow pattern of the capacitated network. The solution of path flows becomes candidate flow patterns, from which we can abstract extreme points specially used for sensitivity analysis, by using a linear programming. Numerical example shows the effect of change in input parameters on flow and delay.