Transportation Geography and Network Science/Telecommunications networks and travel behavior

Introduction
Telecommunication applications such as telephone, Internet and mobile phone have moved the society into the “Information Era” and changed the way people live. The widespread use of telecommunication applications provides significant benefit for both government and individual users. On the other hand, travel patterns are continuously changing throughout human history from carriage to automobile and from railway to the jet plane. As a matter of fact, telecommunication and transportation have some mutual attributes. Both of their purpose is to transport something between origins and destinations. Transportation transports physical objects such as people and goods, while telecommunication transports information by radio waves. Sometimes, the aim of passenger transportation is to conduct some communication at the trip destination. In this sense, transportation sometimes also transports information. As a result, relationships between telecommunication and transportation arose a lot of attention in the past 30 years. The current interest is focusing on two trends: the cost of travel and the functionality of telecommunication in distributing information resources.

Relationships between telecommunication and transportation
Salomon identified four types of relationships between telecommunication and transportation: substitution, complementarity, modification and neutrality. Substitution assumes that the more telecommunication technologies developed and spread, the less will be the demand for travel. On the other hand, complementarity means increase in the use of telecommunication will stimulate travel demand. Substitution and complementarity are two basic types of interactions between telecommunication and transportation systems. There are other two derived relationships: modification and neutrality mean people adjust their travel schedule because of telecommunication and there is no direct relationship between these two systems respectively.

Scale of Study
Research studies focused on the relationship between telecommunication and transportation can be roughly divided into two classes in terms of scale: aggregate (macroscopic) and disaggregate (microscopic) levels. Different empirical studies have different findings that some conclude the potential relationship is substitution while others claim complementarity.

Disaggregated level
For disaggregated level, most studies only focus on specific telecommunication applications such as teleconferencing, telecommuting and teleshopping.

Teleconferencing and telecommuting
Teleconference includes telephone conference, video conference and web conference etc. Popular software for personal use includes Skype, Google Talk, Windows Live Messenger and Yahoo Messenger. The ability of teleconferencing to reduce travel is a very popular notion although research shows that the substitution effect is not obvious. Telecommuting means working at home or other remote location, with connections to a central office. It attracts the most research attention because telecommuting can reduce commuting trips which causes urban congestion. Technically speaking, a great portion of work situations can be easily carried out through telecommunications. However, work still needs face-to-face communications because work also involves social and psychological gratification.

Teleshopping
Conventional shopping requires people to go to the supermarkets, shopping centers and shopping malls, outlets in person. Teleshopping can be considered as non-store shopping where telematics are involved to obtain information or purchase consumer goods. Teleshopping includes shopping channels on cable TV, Internet shopping (not necessary to be goods, but also music, software) From a vendor’s perspective, it is called electronic commerce. However, conventional store shopping still have a lot of attributes that teleshopping doesn’t. For example, the interaction between the customer and salespeople, bargain, opportunity to be outside the home or work environment etc.

Telephone and mobile phone
Telephone and mobile phone are most popular telecommunication applications. Different research found different relationships between travel and telephone/mobile phone use. For example, Yim (1994) studied the role of mobile phone in daily travel in the Bay area. 14.8% in 7347 mobile phone user reported driving less often after getting a cell phone and only 8% reported driving more often. On the other hand, a research done by Yim (2000) 6 years later concluded that mobile phone generated more trips rather than substituting for them.

Although disaggregated approach could provide detailed human behaviors, it is only limited in terms of activities it can capture mainly due to the data for disaggregated research are collected from survey or questionnaires which can’t reflect the whole picture in certain extent (Choo et.al, 2007). In addition, those surveys are restricted to relatively small and homogeneous samples.

Aggregated level
Aggregated level of studies reflects the travel behaviors of larger population over multiple telecommunication applications. It is assumed that they have a bi-directional relationship. Transportation demand has impact on the telecommunication demand. Some telecommunication products and services are directly targeted to transportation use such as vehicle-GPS, and Internet service on planes. On the other hand, telecommunication applications greatly impact travel demand: telecommuting, teleconferencing, teleshopping, telebanking, tele-entertainment etc. It is hypothesized that the transportation infrastructure and telecommunication infrastructure have potential relationships. For example, Internet optical fiber networks are superimposed on existing transportation network such as streets and subway heavily. Embedding the telecommunication infrastructure into the transportation network could reduce the cost of the telecommunication infrastructure. Telecommunication can increase the level of service of a transportation system by improving the efficiency of the network. The obvious example is the Intelligent Transportation System (ITS). The applications of ITS include probe car data collection, sensing technologies (ramp metering, SMART signal), video vehicle detection (Autoscope), HOT (High occupancy toll) lane etc. The relationship between transportation demand/supply and telecommunication demand/supply could be expressed in the following figure:
 * Travel demand vs. telecommunication demand
 * Transportation supply vs. telecommunication supply



In addition to the demand and supply (infrastructure), cost is also related to both demand and supply. Generally speaking, the relationship among demand, supply and cost can be described as a positive feedback loop. That is, when demand increases, the supply will increase which results in a decreasing cost. Then a decreasing cost stimulates the demand again. This trend applies to both telecommunication and transportation system. Economic activity (eg. GPS) and social-demographic (eg. population) also influence all demand, cost and supply in both networks.

Transportation network and telecommunication network
From the network structure and traffic flow point of view, transportation network and telecommunication network also have some similarities.

Flow density relationship
Take computer network as an example. First of all, in computer networks, there exists a similar ‘fundamental diagram’ as in transportation network.



This figure shows the flow-density relationship in both networks. When density is low, flow will increase as density increases in a linear relationship. When density increases above a threshold, the flow will decrease sharply as density continue increases. We call it congestion in both networks when density is high. In transportation network, density can be thought as the number of cars in a certain length of freeway/arterial. In computer network, density can be considered as the number of packets sent by host A to host B within a certain route.

Circuit switching v.s. packet switching
Basically there are two routing techniques in computer network: circuit switching and packet switching. Figure 5 shows how circuit switching works. In circuit switching, a static route is established between the origin (Device A) and destination (Device B) before transferring data. All the demand between A and B will be assigned to this particular route. This technique is similar with the deterministic traffic assignment problem in which we do all or nothing assignment for each O-D pair after the shortest path of each O_D pair is calculated. The advantage of circuit switching is the quality of service is guaranteed for a particular O-D pair. The disadvantage is the static routing may not utilize the network efficiently. On the other hand, packet switching (Figure 6) divided flow between Device A and Device B into small packets. Each packet will choose its route dynamically, meaning when the packet reaches the next node, it will choose its route again based on the current network condition. This feature is very similar to dynamic traffic assignment problems.