Chapter 1 Introduction: intelligent vehicular communications

1.2.1
EVOLUTION OF METHODOLOGY IN TRANSPORTATION RESEARCH
Transportation models generally can be classified into microscopic, mesoscopic, macroscopic, and 
metascopic models. Microscopic models study individual elements of transportation systems, such as 
individual vehicle dynamics and individual traveler behaviors. Mesoscopic models analyze transporta-
tion elements in small groups, within which elements are considered homogeneous. A typical example 
is vehicle platoon dynamics and household-level travel behavior. Macroscopic models deal with ag-
gregated characteristics of transportation elements, such as aggregated traffic flow dynamics and zonal-
level travel demand analysis.
Major research objects in transportation engineering include traffic flow, travel behavior, trans-
portation networks, traffic control and management, freight systems, and other transportation modes. 
The study of traffic flow includes its micro-, meso-, and macroscopic characteristics, human factors, 
autonomous vehicles, and so on. Common approaches include empirical studies, and statistical and 
computer science modeling motivated by new data collection technologies. Theories and models devel-
oped for similar physical objects, such as fluid and particles, are sometimes introduced and improved 
to fit traffic flow characteristics. The research topics of travel behavior include demand analysis, route 

---

6
CHAPTER 1 
INTRODUCTION: INTELLIGENT VEHICULAR COMMUNICATIONS
choice, day-to-day dynamics, and activity choices. Research methods usually involve survey-based 
methods and travel choice models that originated from economics and logistics. Traffic control and 
management involves the design and management of traffic control devices, traveler information provi-
sion, and more recently vehicular communication system. Optimization and control methods are usu-
ally involved. Transportation network consists of traffic flow, traveler behavior, and traffic control. Its 
design and performance evaluation usually rely on integrated models of both planning and operations.
The study of freight systems involves the performance, optimization, and management of commod-
ity flow. Other research objects also include several alternative modes such as public transportation, 
bicycles, and pedestrians, which are important components in transportation systems and can either be 
studied along with behavior model or operational models or together with passenger vehicles as alter-
native studies. These basic research objects remain relatively static throughout the history of transpor-
tation research; however, models to describe and analyze those objects have evolved from generation 
to generation. Meanwhile, technologies play important roles in studying these research objects. More 
detailed data sets can reveal new characters of those objects and lead to new methodologies and models. 
For example, from traditional license-plate matching to inductive loop detectors and to probe vehicle 
technologies, the methodology of estimating and managing traffic flow dynamics on both freeways and 
arterials has evolved from empirical relationship analysis to complicated traffic state estimation and 
advanced traffic control models. Furthermore, similar to the other engineering fields, the evolution of 
transportation models usually involves four major types of contributions: (1) the discovery and intro-
duction of new principles and relationships, (2) the integration of models, (3) the relaxation of ideal as-
sumptions, and (4) performance improvement. The first two types of contributions usually come during 
the transition period between major generations; the second two types of contributions occur regularly 
during all periods. The term model is not used in the contribution because this type of contribution re-
fers only to truly fundamental and original models. Typical examples include the fundamental diagrams 
of traffic flow, kinematic models, and gravity models. One should not underestimate the contribution 
of the latter four types of contributions, since usually the first type of contribution results in raw and
ideal models and formulations that sometimes take years to evolve into practically accurate and 
efficient models that can be applied in the real world, which is quite important for a practical field such 
as transportation. A famous example is the development of the cell transmission model that made solv-
ing the traffic dynamics inferred from LWR model truly efficient and scalable for traffic operations, 
even though it is a category-D contribution. 
Table 1.1
 summarizes the major existing and expected 
contributions and their corresponding types in different generations and different types of models.