A tool that prepares data for an areal interpolation

http://www.mediafire.com/?1s6n2kc891yg4oa

(C) Copyright: Jong-Geun Kim. 2006-2011. All rights reserved.

This tool cannot be used for any commercial purposes without a written permission of the author. Contact the author (j g k i m 2 5 at gmail.com) for any questions.
=======================

This is a ArcGIS tool developed by model builder that prepares data for an areal interpolation. This tool requires two polygon feature classes (FC) and overlays the two. Originally base FC is assumed to be Traffic Analysis Zone (TAZ) while source FC being a census FC (tracts). Source FC must have a field representing population for each feature.

This tool transfers source FC's attributes to base FC while estimating the proportion of population (Pop_Portion) in the overlaid area. This Pop_portion value can be interpreted as the proportion of souce FC's contribution to the base FC. Implicit assumption is that each feature in the source FC has an even distribution of population, which is not necessarily the case with the featuers in the base FC.

Once Pop_portion fied is populated by this tool, next steps are 1) multiply an attribute field with the Pop_portion; 2) summarize the product by the case field uniquely assigned for base FC (e.g: TAZ_Name).

Know the rules of game and be willing to play better within the rules

Publish or perish.

<Stochastic traffic assignment with single-pass algorithm>

The steps of "double-pass" algorithm can be referred to page 288-292 of Sheffi (1984), but we cannot find more details there about the "single-pass" procedure which favor over the original ("double-pass") one.

 

YOSEF SHEFFI (1984) Urban Transportation Networks: Equilibrium Analysis with Mathematical Programming Methods. Prentice-Hall, Inc., Englewood Cliffs, New Jersey 07632

Dial RB. A Probabilistic Multipath Traffic Assignment Model Which Obviates Path Enumeration. Transportation Research 1971;5(2):83-111

Markov Decision Process

BKX (1995) of Transportation Science is about FILM based on probabilistic flows. Since its direct formulation is non-linear, they applied Markov decision process (MDP) to re-formulate the model as a mixed-integer programming.

MDP a discrete time stochastic process. Wiki (http://en.wikipedia.org/wiki/Markov_decision_process) explains it as follows:

At each time step, the process is in some state s, and the decision maker may choose any action as. The process responds at the next time step by randomly moving into a new state s', and giving the decision maker a corresponding reward R_a(s,s'). that is available in state

The probability that the process chooses s' as its new state is influenced by the chosen action. Specifically, it is given by the state transition function P_a(s,s'). Thus, the next state s' depends on the current state s and the decision maker's action a. But given s and a, it is conditionally independent of all previous states and actions; in other words, the state transitions of an MDP possess the Markov property.

Markov decision processes are an extension of Markov chains; the difference is the addition of actions (allowing choice) and rewards (giving motivation). Conversely, if only one action exists for each state and all rewards are zero, a Markov decision process reduces to a Markov chain.

 

This stochastic version of FILM, however, may not be able to consider deviation of consumers at least with current structure of formulation. Because it only looks at the turning probabilities at each node to its incident node and d