The paper establishes an estimation model of urban transportation supply-demand ratio (TSDR) to quantitatively describe the conditions of an urban transport system and to support a theoretical basis for transport policy-making. This TSDR estimation model is supported by the system dynamic principle and the VENSIM (an application that simulates the real system). It was accomplished by long-term observation of eight cities’ transport conditions and by analyzing the estimated results of TSDR from fifteen sets of refined data. The estimated results indicate that an urban TSDR can be classified into four grades representing four transport conditions: “scarce supply,” “short supply,” “supply-demand balance,” and “excess supply.” These results imply that transport policies or measures can be quantified to facilitate the process of ordering and screening them.
This paper describes a methodology for estimating the value of urban transportation supply-demand ratio, TSDR for short, that is the result of interaction between transport system support capacity and inhabitants’ travel requests and can determine whether transport conditions are balanced.
In a multitude of papers, travel demand and transportation supply are shown to act as a means of traffic phenomenon analysis to shed light on the theories and techniques underlying high-effect transportation.
There are several popular theories deriving from supply and demand consideration. For example, kinematic wave theories of lane-changing traffic flow [
A variety of transport problems originating from the contradiction between travel demand and transportation supply have been discussed in the relational literatures. To tackle taxi service refusal [
The theories and techniques mentioned above can be aimed at any part of the transportation system that could make greater contribution. The question of whether transport supply and travel demand influence the whole transportation system has also been discussed.
Travel demand and transportation supply modeling methodology was presented through an Upper-Silesian Conurbation in Poland [
The demand of transportation can be generally defined in terms of inhabitant trips, but the supply aspect had different assumptions according to the object or the aim. While route choice was regarded as a supply aspect of the urban network, the supply curves [
The above literatures are aimed at developing solution or a corresponding theory for a transport problem. However, because the various parts of transportation systems are interactive, the solution of a transport problem is bound to bring up new problems, so this paper proposes a macroscopic analysis method for estimating the TSDR.
A transportation system is an open complex system, and Table
Characteristic of complex system and transportation system.
Complex system [ |
Transportation system characteristic |
|
|
Large numbers of elements are manifold | Elements: |
|
|
Interaction among the elements is more important than the element itself | The crux of transportation system’s maximum efficiency is the coordination of the elements |
|
|
Multiple causality among the elements | Transportation system consists of subsystems such as economy, number of vehicles, environment, travel demand, transport supply, and a traffic congestion subsystem. Every subsystem has causality and there is special causality among the subsystems [ |
|
|
Dynamic and nonlinear | Transportation system’s elements are in a stochastic condition; that is, they vary with time and space; their linear relationship, because of complex causality, cannot satisfy the requirements of modeling to simulate real transportation, so the modeling method has undergone several processes: statistics, differential equations, system dynamic, the models of complex network, and modeling method based on Agent [ |
|
|
Self-organization and self-adaptiveness | A transportation system, because of randomness and complexity, can only operate in orderly fashion by using a self-feedback function. This is, in order for traffic flow tend to be in ordered under certain conditions, the transportation system should self-adjust, based on real-time traffic status, by control and management technologies [ |
System dynamics is an approach to understanding the behavior of complex systems over time, and it is able to deal with internal feedback loops and time delays that affect the behavior of an entire system. This approach was well-suited to strategic issues and could provide a useful tool for supporting policy analysis and decision-making in the transportation field [
This paper analyzes the transportation system using the methods of system dynamics and will estimate the TSDR by VENSIM, an industrial-strength simulation software package for improving the performance of real systems; it has a rich feature set emphasizing model quality, connections to data, flexible distribution, and advanced algorithms.
As stated in
Idealized system of urban transport system.
This is to say that the purpose of measures such as constructing transport facilities, improving services, strengthening management, and pricing reasonably is to offer additional travel service. It is then necessary that transportation supply is quantized by the maximum passenger-carrying capacity of the transport system per unit time. This leads to Definition
The facilities for mass transit in China include urban road, rail transit, and ferry; rail transit mainly includes tram, light rail, rapid rail (metro), monorail, and funicular. Ferry and funicular play an auxiliary role in some cities with special geography conditions such as a river passing through or mountainous composition of the city’s landform; tram generally cannot exist in Chinese cities because of ever more crowded transport. Almost all of the rail transit entities operate independently in underground tunnels or on viaducts so as not to interfere with surface transport, so Hypothesis
There are two transport facilities for metropolitan inhabitant: urban road and rail transit, and both coexist and are independent of one another.
Most often, types of motorized and nonmotorized vehicles operating on urban roads include bus, car, truck, motorcycle, and bicycle; their purpose and travel times vary so traffic composition varies with city limits and times. Because trucks were forbidden in the daytime in most urban districts, cargo traffic’s influence on the transportation supply and travel demand balance can be overlooked in the daytime. Buses, disaggregated by purpose, are composed of public buses, commuter buses, and intercity buses; the latter two are a minority on urban roads and have flexibility for choosing congestion-free routes, so “bus” will refer specifically to public buses to simplify the model. Motorcycles and bicycles are suitable for traveling short distances but rarely run on the expressway and major arterial roads because of the great distances between workplace and home in metropolitan areas. “Car” comes in three forms: private car, official vehicle, and taxi; the first two have similar traits [
Traffic composition on urban roads consists of car, public bus, and taxi; the term “car” includes both private cars and official vehicles.
Travel demand is trip need or expectation of a city inhabitant for business or entertainment and is the natural outgrowth of economic development and urban population increase. It is always reflected in the trip structure of the metropolitan area, combined into several trip modes such as car, bus, taxi, rail transit, bicycle, and walking. The impact factors that affect travel mode choices of urban residents derive from the spatial and temporal nonuniformity of transport facilities’ use, the purpose of people trips, and the convenience of the transport system, so Definition
The trip structure within the metropolitan area can be characterized by four trip modes: car, public bus, taxi, and rail transit. The “car” designation includes both private cars and official vehicles.
This study examined certain cities that exhibited one of the two characteristics listed below. Table The cities are densely populated and relatively well-developed economically so they use superior transportation systems and are able to provide sets of data representing different traffic states for quantitative analysis. Significant transport events occurred in recent years; examples would include the Shenyang metro being in operation and Shenzhen’s new and reformed long urban road. The before-and-after data comparison can reveal an event’s impact on the balance of the transportation system.
Information for chosen cities.
Metropolitan area | Year | Population (million) | Gross Domestic Product (billion, RMB) | Urban area (Km2) | Significant transport event |
---|---|---|---|---|---|
Beijing | 2011 | 20.186 | 1,625.19 | 8579 | |
|
|||||
Tianjin | 2011 | 13.546 | 1,130.73 | 1103 | |
|
|||||
Shanghai | 2011 | 23.475 | 1,919.57 | 9589 | |
|
|||||
Guangzhou | 2012 | 8.223 | 1,355.12 | 2910 | |
|
|||||
Hangzhou | 2013 | 7.253 | 780.20 | 2060 | |
|
|||||
Shenzhen | 2012 | 1.047 | 1,150.55 | 5256 | The 181 km new road and the 133 km reformed road in 2012 |
Shenzhen | 2013 | 1.055 | 1,295.00 | 5282 | |
|
|||||
Shenyang | 2012 | 8.228 | 660.68 | 1504 | Metro line 1 and line 2 opened, respectively, on September 27, 2011, and February 9, 2012 |
Shenyang | 2011 | 7.227 | 591.49 | 1495 | |
Shenyang | 2010 | 7.196 | 501.71 | 1485 | |
|
|||||
Nanjing | 2012 | 8.161 | 720.16 | 1215 |
The sources of the micro- and macrodata involved in this paper are mainly focused on the following ways.
The digital information from
The operational data from the public-transport operation corporation (bus, metro, or taxi operation company) includes not only passengers carried but also Passenger Load Factor, OD (Origin and Destination), and kilometrage of the passengers, used both in directly estimating the supply-demand ratio and reciprocally verifying with official statistics.
Manual records and interview surveys were adopted. The proportion of public buses, taxis, cars and their Average Carried Passengers were manually recorded in forty sections of expressway, major arterial road, minor arterial road, and collector streets during peak hours and off-peak hours during both working days and Sunday. In a roadside interview survey, at least 200 drivers in a city were asked, “Normally, how many kilometers do you drive a day”; the vehicles average travel distance was the average value of these answers.
The transportation supply-demand ratio used to quantitatively describe the transport system state is the ratio between transportation supply and travel demand, that is, between the maximum amount of passengers carried and the total trips according to Definitions
Figure
Urban TSDR estimating framework.
By Hypothesis
Considering that the traffic capacity unit of measurement is the PCU (Passenger Car Unit), the mathematical formulation of bus passengers-carrying supply
Similarly, taxi passengers-carrying supply
The car passenger-carrying supply
The traffic capacity of the urban road net is the maximum number of vehicles running on the urban road net at a certain time; it is limited by the characteristics of the net and the traffic conditions. Figure
Estimation module for traffic capacity of urban road net.
All the transportation modes except rail transit are restricted to the urban road net traffic capacity. And urban road net traffic capacity [
In the above formulas, many parameters may change in implementation of transport policies or measures, so the estimated result of the model can reflect the effect of such changes. For example, investment in transportation may cause length of the roads classified and length of rail transit line increase. By using and generalizing advanced technology like Intelligent Transportation Systems, driver Information System, urban Traffic Area-wide Cooperation Control Systems, and Urban Pedestrian Systems the values of Intersections Effective Utility Coefficient and Lanes Comprehensive Utility Coefficient will increase.
The values of IEUC and LCUC for roads classified are displayed by Time Occupancy [
There are some things to be aware of while observing the IEUC and LCUC: One should try to choose the cross-sections less affected by intersections and the period during which the traffic becomes saturated. The LCUC should be dissected for every lane. Entries to the intersection while measuring the IEUC should be observed. Select all types of the intersections and cross-sections, but not all of them. IEUC and LCUC average aimed at the roads of each grade should be calculated.
Table
Suggested values of IEUC and LCUC.
Urban road classification | Expressway | Major arterial road | Minor arterial road | Collector street |
---|---|---|---|---|
Intersections Effective Utility Coefficient | 0.75 | 0.55~0.65 | 0.45~0.55 | 0.40~0.50 |
Lanes Comprehensive Utility Coefficient | 0.9 | 0.85~0.95 | 0.80~0.90 | 0.85~0.95 |
The rail transit passengers-carrying supply
Estimated module for rail transit passengers-carrying supply.
Generally, the following formulas state the algorithmic method:
those restricted by telecommunication and signal control technology, letting those restricted by the horizontal and vertical curves of rail facilities and the number of the trains. The horizontal and vertical curves determine the train running speeds and the number of the trains supports the Departing Interval. Let
where
In the model shown in Figure
After considering the ratio of the spare and maintaining trains to operating trains (
For the Departing Interval choose the max value between
Beijing subway passengers-carrying estimated verification in 2012.
Rail line | Passengers-carrying estimated (person/hour) | Passengers-carrying in reality (person/hour) | Error rate (%) |
---|---|---|---|
Line 1 | 42336 | 42840 | −1.18 |
Line 2 | 37800 | 38556 | −1.96 |
Line 4 | 26779 | 26677 | 0.38 |
Line 5 | 31129 | 31328 | −0.64 |
Line 8 | 17640 | 17520 | 0.68 |
Line 9 | 11760 | 11680 | 0.68 |
Line 10 | 32072 | 32296 | −0.69 |
Line 13 | 33075 | 32130 | 2.94 |
Line 15 | 12027 | 11680 | 2.97 |
Changping Line | 13569 | 13140 | 3.26 |
Fangshan Line | 9284 | 8760 | 5.98 |
Yizhuang Line | 11386 | 11680 | −2.52 |
Batong Line | 29223 | 29988 | −2.55 |
Airport Express | 2680 | 2688 | −0.30 |
Daxing Line | 22800 | 22603 | 0.87 |
Average error rate (absolute value) | 1.84 |
In Table
The bus lane is one of two types depending on whether it is independent of the urban road net:
In this section, fifteen groups’ data were chosen from eight cities to analyze the TSDR. Table
Estimated result on TSDR.
City | Year | Time period | Bus PCSDR | Rail PCSDR | Taxi PCSDR | Car PCSDR | TSDR |
---|---|---|---|---|---|---|---|
Beijing | 2011 | Peak hour in working day | 0.82 | 0.50 | 0.91 | 0.22 | 0.47 |
Guangzhou | 2010 | Peak hour in working day | 0.92 | 0.37 | 0.95 | 0.56 | 0.68 |
Shanghai | 2011 | Peak hour in working day | 0.81 | 0.57 | 0.93 | 0.60 | 0.69 |
Nanjing | 2012 | Peak hour in working day | 0.86 | 0.84 | 0.90 | 0.66 | 0.76 |
Tianjin | 2011 | Peak hour in working day | 0.83 | 0.96 | 0.86 | 0.76 | 0.81 |
Shenzhen | 2012 | Peak hour in working day | 0.94 | 0.70 | 1.03 | 0.77 | 0.84 |
Shenzhen | 2013 | Peak hour in working day | 0.97 | 0.71 | 0.98 | 0.90 | 0.92 |
Shenyang | 2010 | Peak hour in working day | 0.87 | — | 0.75 | 0.73 | 0.82 |
Shenyang | 2011 | Peak hour in working day | 0.88 | 0.90 | 0.75 | 0.75 | 0.83 |
Shenyang | 2012 | Peak hour in working day | 0.99 | 0.83 | 0.81 | 0.81 | 0.89 |
Hangzhou | 2013 | Peak hour in working day | 0.90 | 1.01 | 0.86 | 0.84 | 0.93 |
Beijing | 2011 | Off-peak hour in working day | 1.13 | 0.96 | 0.85 | 0.93 | 0.97 |
Beijing | 2011 | Weekend daytime | 1.09 | 0.96 | 1.00 | 0.90 | 1.02 |
Shanghai | 2011 | Weekend daytime | 1.07 | 1.12 | 0.96 | 1.05 | 1.06 |
Nanjing | 2012 | Weekend daytime | 1.10 | 1.18 | 1.17 | 1.13 | 1.13 |
Theoretically, the value of supply-demand ratio is near 1: a value greater than 1 means that the supply exceeds the demand, and a value less than 1 means that the supply is short. However, in view of the complexity of the metropolitan transport system and the limitation of the model, the PCSDR of bus, taxi, rail, and car is, respectively, determined, and its underlying causes are explored to lay a foundation for describing the transport conditions represented by the value range of the TSDR.
It is common for the values of bus PCSDR to remain steady during the peak hour interval [
The values of taxi PCSDR are the largest, and they exceed 0.8 in the overwhelming number of major cities. Because the expectation of taxi is the lowest from perspectives of both supply and demand, on the supply side, taxi drivers try to avoid operation to reduce costs, especially under crowded traffic conditions, and on the demand side, passengers often do not choose taxis travel because of the higher trip charge (compared to bus) and the longer trip time (compared to rail transit).
Most of the values of car PCSDR are less than 0.8, especially for Beijing, just 0.22 during peak hours. From the Chinese standpoint, it may be essential to own private cars because families with children, the elderly, and the infirm must use cars as travel tools to avoid dealing with the congestion of urban public traffic. Other reasons for owning a car, a symbol of identity, might include winning more social respect and even bringing about more economic benefits. Therefore, when Beijing and Shanghai use lottery system to limit new vehicle registrations, vehicle possession still had respective increases of 225,000 and 59,000 in 2013 and reached totals of 5.2 and 2.8 million. Relative to urban road mileage, 28,608 and 17,316 kilometers, private car demand greatly exceeds the supply of urban road net, so excessively larger vehicle possession is the root cause of the urban road net congestion. On the other hand, the low average passengers-carrying, only 1.17 from survey data, is another important contributor to congestion.
The estimated values of rail PCSDR range wildly, from 0.37 to 1.01, and the more developed the rail transit system becomes, the lower its value will be. In vast metropolitan subway networks (like those in Beijing, Shanghai, and Guangzhou), during peak hours a horde of people fight their way off the train while another such horde barely waits before fighting their way on, and the train can hardly get moving because of all the people crammed in and blocking doors; subway attendants help by shoving the last people onto the train. In spite of such conditions, more and more people are willing to choose subway travel because this is the only way to arrive at their destination on schedule. It is a kind of inevitable phenomenon that the demand for rail transit in Chinese cities will exceed the supply both now and in the future.
Value range of the TSDR under different transport conditions.
Value range | Significance | Transport conditions |
---|---|---|
|
Scarce supply | Urban transport system cannot meet the challenge of residents travel: excessively crowded rail transit and bus, severe congestion on the road net, and taxi shortage. |
|
||
|
Short supply | In the general case, urban transport system may meet the challenge of residents travel. However, when a sudden event (even a small perturbation) or bad weather is encountered, rail transit will be crowded, many roads will become jammed, and buses will be delayed. In other words, the system has weak ability to withstand disturbance. |
|
||
|
Supply-demand balance | In most situations, urban transport systems can meet the challenge of residents travel and have self-adjustment ability. When a sudden event or bad weather is encountered, some roads will become jammed, and the number of rail and bus passengers will increase. These disturbances will often be quelled in short times without the interposition of managers. |
|
||
|
Excess supply | Under any circumstances the urban transport system can meet the challenge of residents travel. |
In Table
The value of Shenyang’s TSDR also grew from 0.82 in 2010 to 0.89 in 2013, because the Shenyang metro line 1 (operational with 27.8 km and 22 stations) and line 2 (with 27.36 km and 21 stations), respectively, opened on September 27, 2011, and January 9, 2012. The formation of the Shenyang metro network improves the supply capacity of the transport system, but, at the same time, it changed the structure of transportation supply and demand. With more and more people choosing subway travel, the value of TSDR rose, while the values of bus, taxi, and car PCSDR dropped.
In most Chinese cities, it is feasible to think that pushing up transport supply can be adopted to increase the transport system’s efficiency in the near future. However, in metropolitan areas, this approach has lost its foundation because of limited urban space, and reducing travel demand is palliative.
The worst metropolitan transport situation is described in Table
This paper has developed a method for estimating the TSDR and completed the following tasks: (1) the TSDR estimation model was constructed using VENSIM, after idealization based on system dynamic principles, and (2) the estimated TSDR results were analyzed by comparison with fifteen data sets about the eight cities’ transport conditions refined through long-time observation.
The model can provide a basis for transport policy-making because it shows and quantifies the interaction between transport system supply and demand. The TSDR values symbolize the specific transport conditions and a synthetic result of economic, policy, and traffic development. At the same time, the contribution from traffic policies or measures to the TSDR can be evaluated, so investment projections and transport policies or measures can be ordered and screened.
The results of the model will be different for the various selected regions. The TSDR values in the paper reflect the collective transport condition of the cities, but the unequal population density of each region in a city leads to imbalance of their TSDR values. For example, in Guangzhou in 2012 the urban population density was 2060 persons per square kilometer, while values for Yuexiu and Nansha were 15112 and 795, and their TSDR’s values were 0.52 and 0.91 during working-day peak hours, so the geographical scope for the model should be selected according to the particular regional goals for transport policies or measures.
Taken together, this paper sheds light on the nature of likely interaction between transportation supply and demand. However, much work remains to be done because the idealized transport system considered here has a certain distance from reality. Other aspects that clearly deserve further research involve bicycles and motorcycles on the urban road net and changes in the traveling intensity and modes.
The authors declare that there is no conflict of interests regarding the publication of this paper.
This work was partially supported by National Natural Science Foundation of China (Project no. 51278158).