Module route_choice 

Multinomial logit (MNL) route choice model.

Given a set of alternative paths with associated generalized costs, the MNL model selects one path probabilistically:

$$P(k) = \frac{\exp(-\theta \cdot C_k)}{\sum_j \exp(-\theta \cdot C_j)}$$

where:

• $C_k$ is the generalized cost of alternative $k$
• $\theta$ is the scale parameter (larger = more deterministic)

Example

use rustsim_pathfinding::route_choice::mnl_select;
use rand::rngs::StdRng;
use rand::SeedableRng;

let costs = [10.0, 12.0, 15.0];
let mut rng = StdRng::seed_from_u64(42);
let selected = mnl_select(&costs, 1.0, &mut rng);
assert!(selected < costs.len());

References

Ben-Akiva, M. & Lerman, S. (1985). “Discrete Choice Analysis: Theory and Application to Travel Demand.” MIT Press.

Functions

mnl_logsum

Log-sum (inclusive value) of a set of costs under MNL.

mnl_probabilities

Compute MNL choice probabilities for a set of costs.

mnl_select

Select one alternative from a set of costs using the multinomial logit model.