conference paper
published journal article
The authors argue that for the cross-sectional multiattribute approach to choice modeling, the multinomial logit is theoretically and empirically superior to the more commonly used regression approach. Other choice methodologies also are discussed briefly in relation to logit. The difference between individual level (where regression is appropriate) and cross-sectional analysis is recognized. Most marketing managers, because of their research goals, will be using a cross-sectional approach. The derivation of the logit from an underlying behavioral model of choice is illustrated. It is this underlying behavioral model of choice that provides logit with several conceptual advantages for modeling a multiattribute choice structure.
research report
book/book chapter
Housing plays a central role in urbanization and urban sustainability in the United States. Population growth in and migration to urbanized areas is accompanied by demand for housing. The response to this demand, housing development, in turn consumes increasingly more land, spurring urbanization and fostering sprawling development patterns. With over fifty years of growth, mass housing production, and a cultural propensity for consumptive behavior, the United States serves as an exemplary case through which to examine urbanization and the featured role of housing within this larger process. This case also reveals the effects of decisions by housing producers, consumers, and
published journal article
Software-integrated multidomain automotive systems, which is also referred to as automotive cyber-physical systems (CPS) consist of various interacting domains (software, hardware, multiphysics, communication, etc.). Design decisions in one domain may completely change the constraints and requirements in the other domains, e.g., adding more functions in a modern automotive CPS may require changes to thousands of lines of software code or even the mechanical architecture. Existing CPS design methodologies are siloed in a specific domain and therefore have limited design space exploration capabilities because only one domain can be tested at a time. This paper presents a functional-level cyber-physical codesign methodology starting from the functional model of the CPS capable of concurrently expressing (multi-)physics and control in automotive applications. Moreover, we introduce a high-level synthesis algorithm capable of selecting a set of optimized system architectures using various executable simulation components and cost metrics. We demonstrate our methodology with a realistic automotive use case and explore various design alternatives for implementing the control systems in pure continuous domain (e.g., traditional automotive subsystems without engine control units) or hybrid domain (e.g., brake-by-wire, steer-by-wire, drive-by-wire, etc.) under power, performance, and reliability constraints.
research report |
Sep 1993
The research reported herein further develops the DYNASMART simulation program developed at the University of California, Irvine with PATH funding, in informal collaboration with a project funded by the Federal Highway administration at the University of Texas at Austin. The DYNASMART program resulted from the doctoral dissertation of the principal investigator of this PATH project at UT Austin, and has now been developed to include full-scale traffic signal control as well as freeway ramp control. The model is capable of simulating large urban networks under various ATMS and ATIS strategies, and include carefully designed modules for driver responses to information and for capturing the dynamics of the network paths. The model has been implemented on work station as well as mainframe computational platforms. This research has developed a User-friendly front-end for the editing large input files as well as for run-time displays of traffic conditions. The research also included the simulation study of a network in Orange County, California, where the benefits from candidate ATIS and ATMS strategies were evaluated. The DYNASMART model has evolved into a very flexible tool that can be applied to evaluate information and control strategies in realistic urban networks in an efficient manner.
research report |
Sep 1996
working paper |
Nov 1995
