Quantifying a Financially Sustainable Strategy of

Public Transport: Private Capital Investment

Considering Passenger Value
by
Yunqiang Xue 1,2,Hongzhi Guan 1,2,*,Jonathan Corey 3,Heng Wei 3 and Hai Yan 2
1

College of Architecture and Civil Engineering, Beijing University of Technology, Beijing 100124, China
2

Beijing Collaborative Innovation Center for Metropolitan Transportation, Beijing University of Technology,
Beijing 100124, China
3

ART-Engines Transportation Research Lab, University of Cincinnati, Cincinnati, OH 45220, USA

*

Author to whom correspondence should be addressed.

Sustainability 2017, 9(2), 269

Abstract
Releaving traffic congestion by developing public transport as an alternative mode of travel is a common
practice all over the world. However, the increasing public transport subsidies have created a financial
burden for governments. Encouragingly, private capital supplies an opportunity for public transport in
sustainable finance. Previous research mainly focuses on qualitative analysis and money-for-value (MFV)
analysis. In this paper, a new investment model is proposed based on the concept ‘passenger value’, and
a bi-level programming model (BLPM) is constructed as a quantitative analysis tool. The upper target of
BLPM is the total surplus (including the value of time (VOT) of passengers) of the public transport system
and the upper constraint is the ticket price. The lower target of BLPM is passenger’s surplus, the lower
constraints are service capability and the lowest return rate of the private sector. The public transport of
Jinan City, China is taken as a case to quantify the impacts of private capital investment in public
transport. Results show that the proposed investment model considering passenger value is superior to
the traditional one, and effective private capital investment could increase the total societal benefit of the
transportation system. The proposed investment strategy satisfies economic viability and is a financially
sustainability strategy. Additionally, travelers should be encouraged to use public transport through
improving the service quality and passenger returns. Only in this way can the success rate of the private
sector investment in public transport be improved efficiently.
Keywords:
public transport; financial sustainability; passenger value; public-private partnerships; bi-level
programming; travel mode choice

1. Introduction
Traffic congestion, increased travel time, and air pollution are serious problems in modern
metropolitan areas. Easing traffic congestion and solving other related traffic problems through public
transport developments has become a common practice all over the world [1]. In order to guarantee the
basic profit rate of operators, as well as keep the ticket prices at low levels, subsidies are usually needed
for the sustainable development of a public transport system. However, increasing public transport
subsidy requirements have created a financial burden for governments. Financial constraints will affect
bus frequency and bus size, which will affect the service quality of public transport [2]. Therefore,
governments at all levels have to face the problem of maintaining service quality and keeping the transit
fares at low levels based on limited financial resources [3]. How to solve the problem is worthy of
research and discussion.
 Encouragingly, private capital provides a potential solution to the above problem because the private
sector can contribute in terms of financial support, technical skills, innovation, technology advances,
specialist knowledge, and efficiency [4]. The experiences of developed countries show that the way out of
the dilemma for public transport is to establish competition between public and private sectors, and
introduce private capital in the urban public transport industry [5,6,7,8,9,10,11,12,13]. After 2000, more
and more countries became interested in introducing private capital to the fields of construction and
operations in public affairs, and Public-Private Partnership (PPP) is a common method. However, there is
no single, internationally-accepted definition of PPP. The definition given by the World Bank group—a
long-term contract between a private party and a government entity, for providing a public asset or
service, in which the private party bears significant risk and management responsibility, and remuneration
is linked to performance [14]—will be used here. Many EU countries have a powerful incentive to explore
alternative funding approaches to build transport infrastructure and provide service delivery. Most EU
member states and the European Commission regard PPP as an important method to attract additional
financial resources for high-priority investments, such as transportation [9,15]. Many states in America
have a great interest in Public-Private Partnerships due to fiscal constraints, declining infrastructure, and
increased industry specialization, e.g., in the early 1990s Texas, Virginia, Colorado, and Florida added
new road and rail capacity to their most congested corridors with PPP concessions financed with private
and public funds [16]. In developing countries, private sectors are also introduced to the procurement of
public facilities and services [17,18,19].
Chinese governments at all levels encourage private capital investment in public transport via
legislative and regulatory reforms in order to ease the financial pressure [17,20,21,22,23]. However, the
investments in transportation operations have largely failed due to unsustainable financial operations and
some other factors (e.g., lack of well-established legal framework, poor governance, political/social
obstacles, and so on.) [17,24,25,26,27,28]. Zhang [24,25,26,27] divides PPP failures into six types: value
for money is not achieved, concession cancelled, concession tender cancelled, project nationalization,
project halted, and contract suspension. Financial unsustainability is a key failure factor. After the 1990s,
the public transport industry in China experienced a tide of marketization. Due to the low level ticket price
and subsidies restricting the private sectors’ profit rates, most of the investments in public transport
operations eventually failed. Public transport returned to be state-owned. Due to financial burdens,
Chinese governments encouraged private capital investment in public transport again after 2010. Under
this background, improving the success rate and attracting private capital investment in public transport is
a problem for governments and researchers [4] (a PPP project is successful only if the contract items are
realized, the success rate means the ratio of successful PPP projects over total PPP projects in a specific
time period).
There have been many research projects focused on the critical success factors (CSF) for
introducing private capital into public affairs [12]. Osei-Kyei and Chan [12] reviewed the studies of CSFs
for implementing PPP from 27 top-tier academic journals from 1990–2013. The most identified success
factors are risk allocation and sharing, strong private consortium, political support, community/public
support, and transparent procurement [12]. Chan [4] researched privileges and attractions for the private
sector to involve itself with PPP projects in mainland China and Hong Kong. Three attractions and
privileges were rated higher by respondents in mainland China: government guarantee (joint first
position), government assistance in financing (joint first position), and tax exemption/reduction. Two
factors observed to be higher for Hong Kong respondents were government sponsorship and incentives
for new market penetration [4]. The above research results provide guides for introducing public private
investments in public transport.
It is obvious that if the feasibility of private capital investment in public transport can be quantitatively
evaluated, then decision-makers (government officials in management departments of public transport)
can make logical investment policies more easily [16,29,30,31,32,33,34,35]. However, many studies
focus on qualitative analysis [12,17,28,36]. The quantitative studies mainly focus on value for money
(VFM) analysis of infrastructure construction and cost-benefit analysis (CBA) [17,31,32,33,34,35,36,37].
The concept VFM may differ between agencies. VFM is often regarded as a monetary value savings
produced by selecting the PPP option versus a traditional procurement option. If the value is positive, the
project should be chosen. Otherwise, the project should be dropped [37]. The definition given by the UK’s
HM Treasury is ‘optimum combination of whole-of-life’ costs and quality (or fitness for purpose) of the
goods or service to meet the user’s requirement [38]. Current VFM analysis mainly deals with the
economic aspect of ensuring VFM in PPP projects. Other non-financial costs and benefits are neglected
in the quantitative assessment [38]. Factually, the value-of-time (VOT) of passengers is also important
due to public necessity of transport [3]. VOT should be considered because the travel time is different
before and after providing public transport services. The saved time can be measured in a monetary form
as a part of the returns of government subsidies in public transport [39]. Additionally, passengers form a
vast customer base which can generate huge values, beyond ticket fares. However, the customer value
of passengers (‘passenger value’ for short in the following sections, and the concept is explained in next
section) is not fully considered when private capital invests in public transport. In the UK, and many other
European countries, private sectors are bidding for subsidy payments from the public authority through
providing public transport service passenger services [40], because the user charges are not enough for
private sectors to achieve financial sustainability [9]. However, the private sectors did not utilize the
resources of public transport passengers fully, with respect to user charges. Otherwise, traveler’s choice
behavior should also be considered, which will affect the number of passengers. Since there is
competition between different traffic modes, travelers may transfer among travel modes due to the
generalized costs (travel time, travel cost, service quality, and so on) [41].
As mentioned above, passengers form a customer base that can generate other values beyond
ticket fares. If the passenger value could be fully considered by private sectors, it would be valuable for
financial benefits. To address the shortcomings of previous research on private capital investment in
public transport, namely, a lack of passenger value analysis and traveler’s mode choice behavior, not
considering the VOT of passengers, a new model of private capital investment in public transport
considering passenger value was developed. A bi-level programming model (BLPM) was constructed as
a tool of quantifying the impacts of the sustainable development strategy for public transport, and the
lower level function took traveler’s mode choice behavior into account. The model validity was verified
based on a case study of Jinan city traffic in China. Conclusions, policy recommendations, and further
research issues are provided in the last section.

2. Private Capital Investment Models in the Transportation Industry

In this research, the definition of private capital in [3] is adopted; private capital means all domestic
capital excluding state/government investment capital. Private capital investments in the transportation
industry mainly concentrate in three areas: transportation services, infrastructure construction, and the
emerging transportation industry [42]. PPP is an important method to introduce the private sector to
transportation. However, PPP is different from privatization. In the privatization processes, there is a
definitive transfer of the ownership of the assets [43]. Xu [44] concluded the forms of PPP and degrees of
private involvement in Figure 1. Percoco [45] and the World Bank provide the descriptions of PPP types
(Table 1).
Figure 1. Forms of Public-Private Partnership (PPP) and degrees of private involvement [44].
Table 1. Description of PPP types [45].

Percoco [45] classified all PPP contracts into 12 types (Table 1) and grouped them into four
categories (management and lease contracts, concessions, greenfield projects, and divestitures) which
can be ranked by increasing private participation and risk transfer from public sectors to private ones.
Private capital investment in urban public transport is limited in scope [3] to mainly infrastructure
construction and operations. Since investments in construction are well-addressed, investments in
operations are particularly discussed in the following sections.

3. The Investment Model in Public Transport Considering Passenger Value

Seybold, et al. [46] stated that the ‘customer economy’ era had come. Research on customer value
is based on three different approaches [46,47,48,49]: (1) Customers are treated as the subjects and
enterprises as objects. To understand and meet customer demand is the major task; (2) Customers are
regarded as objects and enterprises as subjects. The primary task is attracting and maintaining profitable
customers; and (3) Customers and enterprises are regarded as both subjects and objects, known as a
customer value exchange. The customer value exchange means customers can benefit from enterprises
and, at the same time, enterprises can benefit from customers. It considers benefits of both customers
and enterprises, while approach (1) or (2) mainly consider the benefits of customers or enterprises. In
public transport systems, the customer value of passengers (‘passenger value’ for short) has two
meanings: For public transport operators, passenger value means the benefits brought by passengers,
including the fare income, cash flow income of the public transport card balance, and advertisement
revenue; the other aspect of passenger value is the travel service provided to passengers by public
transport companies. The third approach is widely used at present, and this paper will use this approach
in the following analysis.
Customer value analysis is widely used in the areas of mobile commerce and telecommunications
[50,51]. Research in the transportation industry is rare, mainly focusing on railway customer value
classification [52,53]. However, the equivalent research about passenger value in the public transport
industry is lacking.
Passengers form a vast customer base which can generate value besides ticket fares. The purpose
of considering passenger value in public transport system is to increase the system welfare. The system
welfares include the surpluses of all of the groups in the public transport system. There are two major
surpluses in mainstream economics: consumer surplus and producer surplus. Consumer surplus is the
monetary gain obtained by consumers because they want to purchase a product or some service for a
price that is less than the highest price they are willing to pay. Producer surplus is the benefit gained by
selling a product at a market price that is higher than the lowest price that producers would be willing to
sell for [3]. Herein, the public transport enterprise surplus equals the benefit minus cost, and the traveler
surplus equals the fee public transport travelers are willing to pay minus the actual cost. If the passenger
value is fully considered, it should be beneficial for both public transport companies and passengers. After
considering passenger value fully, public transport companies can gain more profits besides ticket fees,
e.g., more profits of cash flow in accounts of public transport cards (Table 2), because the profits of cash
flow in accounts of public transport cards can change from bank interest to investment profit [3]. In many
countries there is no current interest, the current interest rate in China is about 0.3%. On the contrast, the
return on investment is generally larger than 10%. Passengers’ surplus can also increase due to the
returns to passengers (Table 2). Thus, it should be more conducive to sustainable private capital
investment in public transport operation, since some investments have failed due to unsustainable
financial operations.
Table 2. Differences in profits with and without considering passenger value.

A lease is taken as an example, and other types can be analyzed similarly. The government leases
the public transport route to a private operator for a fee. The private operator takes on the operational risk
(see Table 1). Passenger value and service quality constraints are considered in the lease contract, and
the private operator is permitted to utilize passenger resources as a customer base and make use of cash
flow in accounts of passengers’ public transport cards. Then, the private sector can gain other profits
besides ticket fares [3]. The private operator also shares the same government subsidies as the
traditional public transport company. Based on the report “White Book of Jinan Urban Public Traffic
Development” [54], government subsidies include fuel allowances, vehicle subsidies, trips allowances for
special social groups (students, the aged, the disabled, soldiers), and other subsidies for policy-related
losses. Ticket price ranges and minimum service levels are listed in the contract. When the public
transport system efficiency is measured, the value-of-time (VOT) of passengers [39] is considered as a
basis for fiscal subsidy since the government purchases public services to supply travel services
[3]. Table 2 shows the differences in profits for the private sector with and without considering passenger
value. One difference is the profits of cash flow in the accounts of passengers’ public transport cards
change from bank interest to investment profit. It is obvious that the rate of investment profit is higher than
that of bank interest (the current interest of a bank account in China is about 0.3%, while there is no
current interest in many Western countries (e.g., USA). It is common the investment profit rate is higher
than that of bank interest). Another difference is that advertising profit is increased by e-commerce ads.
The private sector may give some profits to passengers who take more trips by public transport. Then,
different customer values of passengers can be reflected and more people will be enticed to travel by
public transport. According to the financial statement of the Jinan Public Traffic Company, the fixed cost
contains office expenses, property costs, vehicle insurance, and depreciation charges. The variable cost
contains consumptions of fuels, materials, and lubricants, as well as repair charges, accident costs,
salaries, and all kinds of insurances.
In addition to following a sound financial system and regulations, private sectors should improve
efficiency of resource utilization, since private sectors can gain more profits and enhance competitive
power if they make use of passenger resources more efficiently. More profits are what the private sectors
want. The investment model proposed here considers passenger value, generates more profits, and
should be more conducive to private capital investment in public transport. In the following section, a
quantitative tool is constructed to measure the specific impacts of private capital investment in public
transport considering passenger value.

4. BLPM of Private Capital Investment in Public Transport

Xue and Guan [3] also established the BLPM of private investment in public transport in previous
research. This paper is an extended research of the previous study. The major difference is that private
cars and taxis were added to the competition among travel modes. Private cars and taxis were also
considered in the case study. Additionally, there are more parameter analysis and policy
recommendations in this paper than the previous one, and the former paper mainly focused on model
characteristics. Although the model structures are similar, the model contents are obviously different.
Tables and figures are significantly different. The data is also different in the two papers.
The bi-level programming problem (BLPP) is a programming model which contains upper and lower

programming [55]. Hypothesizing that the control variable of the upper level function is 𝑥=(𝑥1,𝑥2,
levels. Additionally, the lower level programming is the constraint condition for the upper level

⋯,𝑥𝑛)𝑇∈𝑋⊂𝑅𝑛. The control variable of the lower level function is 𝑦=(𝑦1,𝑦2,⋯,𝑦𝑛)𝑇∈𝑌⊂𝑅𝑚.

𝑃1
Therefore, the BLPM is listed as follows:
min𝑥∈𝑋𝐹(𝑥,𝑦)
(1)
𝑠.𝑡.𝐺(𝑥,𝑦)≤0, 𝑦 is solved from 𝑃2
(2)
𝑃2 min𝑦∈𝑌𝑓(𝑥,𝑦)
(3)
𝑠.𝑡.𝑔(𝑥,𝑦)≤0
(4)
where, functions 𝐹,𝑓:𝑅𝑛1×𝑅𝑛2→𝑅 are the respective target functions of P1 and P2; Vector
functions 𝐺:𝑅𝑛1×𝑅𝑛2→𝑅𝑚1,𝑔:𝑅𝑛1×𝑅𝑛2→𝑅𝑚2 are respective constraint conditions of P1 and P2.
 Figure 2 shows why the multi-level programming theory is used to establish a quantitative
evaluation model for optimization analysis of the effects of private capital investment in public transport.
There are three groups in a public transport system: government officials, public transport companies,
and travelers. Each group is at a different hierarchical level. Government requires the public transport
system generate the largest societal benefit, while public transport companies want to get the maximum
profit, and travelers hope to minimize trip cost. Public transport companies make operation strategies
based on the service quality constraint. Travelers will choose the suitable traffic modes according to travel
choice behavior theory. The travel choices will, in return, affect decision-makers (government officials). If
fewer travelers choose public transport, government officials have to change strategies (e.g., subsidies to
private sectors, lease fees of private sectors, reset the range of ticket fares and the constraint of the
service level of public transport) to improve the service quality. Then, facing the new strategy of the
decision-maker, operators should adjust operational strategies (e.g., ticket fares, departure frequencies,
the rate of returns to passengers, and advertising business). The iterative process will not stop until the
public transport system is optimal. In fact, if the system efficiency is acceptable by government officials,
the system is stable and the present situation will be maintained. The relationships and restrictions can be
captured effectively by multilevel programming models. Since public transport belongs to public affairs,
the government purchases public services to supply travel services. Therefore, governments do not
expect public transport companies to maximize profits, although they have to meet the minimum income
constraint of public transport companies [3,56]. The three-level decision problem can be simplified into the
BLPP. Since the system welfare contains not only money but also some non-financial benefits (e.g., travel
time), the generalized trip cost is considered in the BLPM.

Figure 2. The strategy iterative flow of hierarchical groups in public transport system optimization.
Suppose there are three basic kinds of public transport modes: bus, bus rapid transit (BRT), and
subway (there are also light rail train, electric tram, trolley bus, and maglev train in some cities; in the
following analyses we only consider bus, BRT, and subway for simplicity. The public transport modes are
treated as the subscript values of the parameters. Only the subscripts take different values if other modes
are considered; actually, this will not change the model structure). Suppose the private sector invests in
one of the public transport modes and pays some lease fee to the government. The private operator also
shares the same government subsidies as the traditional bus company. The government sets the ticket
price ranges and requires minimum service levels. The ticket price ranges are based on the per capita
level (the average spending on public transport lies in the range of 5%–10% of personal disposable
income) and individual passengers’ acceptable spending level [3]. The upper decision-makers
(government officials in the management department of public transport) aim to realize the largest social
welfare of the public transport system (including traditional public transport companies’ surplus, private
sector lease fee, and traveler’s surplus) [3]. The upper control variable is the ticket price. Travelers in the
lower programming model aim to minimize the generalized travel cost, and the lower control variable is
the service level (measured by departure frequency). The constraint conditions are the profit rate of the
private sector and the vehicle capacity limit. Although more travelers choosing public transport will be
beneficial to transportation system, travelers are not compelled to give up private cars. Travelers choose
their desired travel mode through choice theory. In order to show the impacts of car usage to the welfare
of the public transport system, the surplus of car travelers is considered in bi-level objective functions.
Normally, the generalized cost of travel by car is higher than by public transport, and it is a common
practice to encourage citizens to travel by public transport for transportation system optimization.
The BLPM of private capital investment in public transport operation considering passenger value is
presented as follows:
max𝑓𝑆(𝑓,𝐹)=∑𝑠∑𝑡∫0𝐷𝑠𝑡𝑓𝑠𝑡(𝑑)𝑑𝑤+𝐵𝑝𝑟𝑖𝑣𝑎𝑡𝑒+𝐶𝑝𝑟𝑖𝑣𝑎𝑡𝑒0(𝑄𝑝𝑟𝑖𝑣𝑎𝑡𝑒)
+∑𝑖=1𝑛𝐵𝑠𝑢𝑏𝑠𝑖𝑑𝑦𝑖+∑𝑖=1𝑛𝐵𝑜𝑡ℎ𝑒𝑟𝑖−∑𝑖=𝑛+1𝑚𝑓𝑖𝑄𝑖−∑𝑠∑𝑡𝑓𝑐𝑎𝑟𝑠𝑡𝑄𝑐𝑎𝑟𝑠𝑡−𝐶𝑤𝑖𝑡ℎ𝑜𝑢𝑡0(𝐹)
(5)
𝑠.𝑡. 𝑓𝑖min≤𝑓𝑖≤𝑓𝑖max
(6)
𝑟𝑏𝑎𝑛𝑘≤𝑟𝑝𝑟𝑖𝑣𝑎𝑡𝑒≤∑𝑖=𝑛+1𝑚𝑓𝑖𝑄𝑖+𝜑(𝑇,𝑄𝑝𝑟𝑖𝑣𝑎𝑡𝑒,𝑀𝑎𝑐𝑜𝑢𝑛𝑡)
+∑𝑖=𝑛+1𝑚𝐵𝑠𝑢𝑏𝑠𝑖𝑑𝑦𝑖+∑𝑖=𝑛+1𝑚𝐵𝑜𝑡ℎ𝑒𝑟𝑠−𝐶𝑝𝑟𝑖𝑣𝑎𝑡𝑒0(𝐹)−𝐶𝑝𝑟𝑖𝑣𝑎𝑡𝑒0(𝑄𝑝𝑟𝑖𝑣𝑎𝑡𝑒)
−𝐵𝑝𝑟𝑖𝑣𝑎𝑡𝑒𝐶𝑝𝑟𝑖𝑣𝑎𝑡𝑒0(𝐹)+𝐶𝑝𝑟𝑖𝑣𝑎𝑡𝑒0(𝑄𝑝𝑟𝑖𝑣𝑎𝑡𝑒)+𝐵𝑝𝑟𝑖𝑣𝑎𝑡𝑒
(7)
min𝐹𝑇𝐶(𝑆)=∑𝑖𝑄𝑖𝐶𝑖𝑡−𝐶𝑝𝑟𝑖𝑣𝑎𝑡𝑒0(𝑄𝑝𝑟𝑖𝑣𝑎𝑡𝑒)+∑𝑠∑𝑡𝐶𝑐𝑎𝑟𝑠𝑡𝑄𝑐𝑎𝑟𝑠𝑡
(8)
𝑠.𝑡. 𝐹𝑖min≤𝐹𝑖≤𝑁𝑇𝑖(2𝑟𝑡𝑖)
(9)
𝑞𝑖≤𝑇𝐶𝑎𝑝𝑖
(10)
𝑄𝑝𝑢𝑏𝑙𝑖𝑐=∑𝑖=1𝑚𝑄𝑖*(𝑓𝑖,𝐹𝑖)
(11)
Equation (5) is the upper objective function, which represents the welfare of the public transport

sector 𝐵𝑝𝑟𝑖𝑣𝑎𝑡𝑒, and traveler’s surplus (Equation (13)). 𝐵𝑝𝑟𝑖𝑣𝑎𝑡𝑒 is a parameter, its value is
system. It is the sum of traditional bus company’s surplus (Equation (12)), lease fee of the private

determined in the contract. The value constraint of 𝐵𝑝𝑟𝑖𝑣𝑎𝑡𝑒 can be determined in the model; too high
a lease fee will affect the profit rate of the private sector, while too low a lease fee is not what decision-

𝑆𝑝𝑢𝑏𝑙𝑖𝑐𝑡𝑟𝑎𝑛𝑠𝑝𝑜𝑟𝑡0(𝑓,𝐹)=∑𝑖=1𝑛𝑓𝑖𝑄𝑖+∑𝑖=1𝑛𝐵𝑠𝑢𝑏𝑠𝑖𝑑𝑦𝑖+∑𝑖=1𝑛𝐵𝑜𝑡ℎ𝑒𝑟𝑖−𝐶𝑤𝑖𝑡ℎ𝑜𝑢𝑡0(𝐹)
makers want. Public transport operator’s surplus and traveler’s surplus are given as follows:

(12)
𝑆𝑡(𝑓,𝐹)=∑𝑠∑𝑡∫0𝐷𝑠𝑡𝑓𝑖𝑗(𝑑)𝑑𝑤−∑𝑖=1𝑚𝑓𝑖𝑄𝑖+𝐶𝑝𝑟𝑖𝑣𝑎𝑡𝑒0(𝑄𝑝𝑟𝑖𝑣𝑎𝑡𝑒)−∑𝑠∑𝑡𝑓𝑐𝑎𝑟𝑠𝑡𝑄𝑐𝑎𝑟𝑠𝑡
(13)
max𝑓𝑆(𝑓,𝐹)=𝑆𝑝𝑢𝑏𝑙𝑖𝑐𝑡𝑟𝑎𝑛𝑠𝑝𝑜𝑟𝑡0(𝑓,𝐹)+𝐵𝑝𝑟𝑖𝑣𝑎𝑡𝑒+𝑆𝑡(𝑓,𝐹)
(14)
where:
 𝑆𝑝𝑢𝑏𝑙𝑖𝑐𝑡𝑟𝑎𝑛𝑠𝑝𝑜𝑟𝑡0: public transport operator’s surplus;
𝑓𝑖: ticket price of route i;


𝑄𝑖: passenger flow of route i;



𝑄𝑝𝑢𝑏𝑙𝑖𝑐: total passenger flow of public transport;



𝐵𝑠𝑢𝑏𝑠𝑖𝑑𝑦𝑖: subsidy of route i;



𝐵𝑜𝑡ℎ𝑒𝑟𝑖: other benefit of route i;



𝐶𝑤𝑖𝑡ℎ𝑜𝑢𝑡0(𝐹): the operation cost without private capital investment under




𝑆𝑡: surplus of travelers, equal to the cost bus travelers are willing to pay for the
departure frequency F;

service minus the actual cost, minus the travel cost of car users. This means the fewer car

𝐷𝑠𝑡: public transport travel demand under O-D (origin-destination) pair s and t,
users the better, although it is not compulsory for travelers to drop private cars.


𝐷=𝐷(𝑓,𝐹);
which is a function of ticket price and bus service;

𝑓(𝑑): the inverse function of public transport demand, reflects the corresponding



𝐶𝑝𝑟𝑖𝑣𝑎𝑡𝑒0(𝑄𝑝𝑟𝑖𝑣𝑎𝑡𝑒): returns to passengers ‘Q’ by the private capital enterprise;

price and service corresponding to demand D;

𝑓𝑐𝑎𝑟𝑠𝑡: the travel cost of private car from origination s to destination t;



𝑄𝑐𝑎𝑟𝑠𝑡: the travel flow of private car from origination s to destination t;



 m: the number of total public transit lines;
 n: the number of state-owned lines, n ≤ m;
The constraint condition given by Equation (6) is the compulsive ticket price range of route i set by
the government.

Where, 𝑟𝑏𝑎𝑛𝑘 is the rate of bank interest and 𝑟𝑝𝑟𝑖𝑣𝑎𝑡𝑒 is the minimum profit rate the private sector
The constraint condition given by Equation (7) is the revenue constraint for the private sector.

can accept. 𝜑 is a beneficial function, and is related with tine ‘T’, passenger flow ‘Q’, and bus card

‘T’. 𝐶𝑝𝑟𝑖𝑣𝑎𝑡𝑒0(𝐹) is the operation cost of private sector under departure frequency F.
account balance ‘Maccount’. In the case study, it is presented as the product of ‘Q’ and ‘M’ under a fixed time

The lower objective function given by Equation (8) is the generalized travel cost of travelers. The
more detailed function is listed as follows:
min𝐹𝑇𝐶(𝑆)=∑𝑖𝑄𝑖𝐶𝑖𝑡−𝐶𝑝𝑟𝑖𝑣𝑎𝑡𝑒0(𝑄𝑝𝑟𝑖𝑣𝑎𝑡𝑒)+∑𝑠∑𝑡𝑄𝑐𝑎𝑟𝑠𝑡𝐶𝑐𝑎𝑟𝑠𝑡 =∑𝑖𝑓𝑖𝑄𝑖+
[𝛽𝑏𝑢𝑠𝑡𝑟𝑎𝑣𝑒𝑙(∑𝑄𝑏𝑢𝑠𝑇𝑏𝑢𝑠𝑡𝑟𝑎𝑣𝑒𝑙)+𝛽𝑏𝑢𝑠𝑤𝑎𝑖𝑡(∑𝑄𝑏𝑢𝑠𝑇𝑏𝑢𝑠𝑤𝑎𝑖𝑡)+𝛽𝑏𝑢𝑠𝑤𝑎𝑙𝑘(∑𝑄𝑏𝑢𝑠𝑇𝑏𝑢𝑠𝑤𝑎𝑙𝑘)] +
[𝛽𝑏𝑟𝑡𝑡𝑟𝑎𝑣𝑒𝑙(∑𝑄𝑏𝑟𝑡𝑇𝑏𝑟𝑡𝑡𝑟𝑎𝑣𝑒𝑙)+𝛽𝑏𝑟𝑡𝑤𝑎𝑖𝑡(∑𝑄𝑏𝑟𝑡𝑇𝑏𝑟𝑡𝑤𝑎𝑖𝑡)+𝛽𝑏𝑟𝑡𝑤𝑎𝑙𝑘(∑𝑄𝑏𝑟𝑡𝑇𝑏𝑟𝑡𝑤𝑎𝑙𝑘)]
−𝐶𝑝𝑟𝑖𝑣𝑎𝑡𝑒0(𝑄𝑝𝑟𝑖𝑣𝑎𝑡𝑒) +[𝛽𝑠𝑢𝑏𝑤𝑎𝑦𝑡𝑟𝑎𝑣𝑒𝑙(∑𝑄𝑠𝑢𝑏𝑤𝑎𝑦𝑇𝑠𝑢𝑏𝑤𝑎𝑦𝑡𝑟𝑎𝑣𝑒𝑙)
+𝛽𝑠𝑢𝑏𝑤𝑎𝑦𝑤𝑎𝑖𝑡(∑𝑄𝑠𝑢𝑏𝑤𝑎𝑦𝑇𝑠𝑢𝑏𝑤𝑎𝑦𝑤𝑎𝑖𝑡)+𝛽𝑠𝑢𝑏𝑤𝑎𝑦𝑤𝑎𝑙𝑘(∑𝑄𝑠𝑢𝑏𝑤𝑎𝑦𝑇𝑠𝑢𝑏𝑤𝑎𝑦𝑤𝑎𝑙𝑘)]+
[∑𝑠∑𝑡𝑓𝑐𝑎𝑟𝑠𝑡𝑄𝑐𝑎𝑟𝑠𝑡+𝛽𝑐𝑎𝑟𝑡𝑟𝑎𝑣𝑒𝑙(∑𝑠∑𝑡𝑄𝑐𝑎𝑟𝑠𝑡𝑇𝑐𝑎𝑟𝑡𝑟𝑎𝑣𝑒𝑙𝑠𝑡)].
(15)
 𝑇𝐶(𝑆): trip cost of travelers under private capital investment type S (in the

𝑄𝑖: volume of route i;

paper S means lease);

𝑄𝑐𝑎𝑟𝑠𝑡: volume of private car from origination s to destination t;



𝐶𝑖𝑡: generalized travel cost of route i;



𝐶𝑐𝑎𝑟𝑠𝑡: generalized travel cost of private car from origination s to destination t;



𝛽𝑏𝑢𝑠𝑡𝑟𝑎𝑣𝑒𝑙: travel time value of the bus (yuan/hour);



𝑇𝑏𝑢𝑠𝑡𝑟𝑎𝑣𝑒𝑙: travel time by bus;



𝛽𝑏𝑢𝑠𝑤𝑎𝑖𝑡: wait time value of the bus (yuan/hour);



𝑇𝑏𝑢𝑠𝑤𝑎𝑖𝑡: wait time of the bus;



𝛽𝑏𝑢𝑠𝑤𝑎𝑙𝑘: walk time value of the bus (yuan/hour);



𝑇𝑏𝑢𝑠𝑤𝑎𝑙𝑘: walk time of the bus;



𝛽𝑏𝑟𝑡𝑡𝑟𝑎𝑣𝑒𝑙: travel time value of the BRT (yuan/hour);



𝑇𝑏𝑟𝑡𝑡𝑟𝑎𝑣𝑒𝑙: travel time of the BRT;



𝛽𝑏𝑟𝑡𝑤𝑎𝑖𝑡: wait time value of the BRT (yuan/hour);



𝑇𝑏𝑟𝑡𝑤𝑎𝑖𝑡: wait time of the BRT



 𝛽𝑏𝑟𝑡𝑤𝑎𝑙𝑘: walk time value of the BRT (yuan/hour);
𝑇𝑏𝑟𝑡𝑤𝑎𝑙𝑘: walk time of the BRT;


𝛽𝑠𝑢𝑏𝑤𝑎𝑦𝑡𝑟𝑎𝑣𝑒𝑙: travel time value of the subway (yuan/hour);



𝑇𝑠𝑢𝑏𝑤𝑎𝑦𝑡𝑟𝑎𝑣𝑒𝑙: travel time of the subway;



𝛽𝑠𝑢𝑏𝑤𝑎𝑦𝑤𝑎𝑖𝑡: wait time value of the subway (yuan/hour);



𝑇𝑠𝑢𝑏𝑤𝑎𝑦𝑤𝑎𝑖𝑡: wait time of the subway;



𝛽𝑠𝑢𝑏𝑤𝑎𝑦𝑤𝑎𝑙𝑘: walk time value of the subway;



𝑇𝑠𝑢𝑏𝑤𝑎𝑦𝑤𝑎𝑙𝑘: walk time of the subway;



𝛽𝑐𝑎𝑟𝑡𝑟𝑎𝑣𝑒𝑙: travel time value of private car users;



𝑇𝑐𝑎𝑟𝑡𝑟𝑎𝑣𝑒𝑙𝑠𝑡: travel time of private car from s to t.



The inequality shown in Equation (9) is the departure frequency constraint. NTi is the largest number
of vehicles in route i. rti is the average travel time of rout i (hour) in one running cycle.
The inequality shown in Equation (10) is the vehicle capacity constraint, where, TCapi is the average
capacity of vehicles on route i.
Equation (11) is the traffic demand constraint under travel choice behavior theory. The multinomial
logit model is used to describe travelers’ mode choice behavior.
The multinomial logit model (MNLM) is the basic disaggregate model and is mainly used for
describing choice behaviors [57,58]. There are also other choice behavior theories, including game
theory, prospect theory, and theory of planned behavior (TPB). Since the disaggregate theory is robust
and widely used at present [59,60,61], it will be used in the paper. The disaggregate model is based on
the hypothesis that travelers choose the maximal utility option under a specific condition. The probability

𝑃𝑖𝑛=exp(𝜃𝑉𝑖𝑛)∑𝑖=1𝐽exp(𝜃𝑉𝑖𝑛), 𝑖=1,2,⋯,𝐽
that traffic mode i will be chosen by traveler n is presented as follows:

(16)
where, 𝑉𝑖𝑛 is the deterministic component of utility Ui for traveler n; 𝜃 is a parameter; and J is a set
of available traffic mode alternatives for travelers. The parameter 𝜃 can be estimated through maximum
likelihood and the Newton-Raphson method [57].

5. Solution Algorithm
Due to the NP-hard problem, it is difficult to obtain the global optimal value. Colson published a
series of representative research results to solve the solution problems of nonlinear bi-level programming
models [62,63,64]. There are two types of methods to solve the bi-level programming problem: one
method is to transform the bi-level programming model into a single programming model based on the
optimal conditions, such as KKT conditions; the other method is to obtain the optimal solution of the lower
model under a given variable value of the upper model. Based on the two types of solution methods,
many specific algorithms have been proposed and developed [64,65,66,67,68,69].
Intelligent algorithms, such as genetic algorithms and simulated annealing algorithms, have a wide
range of applications, because these algorithms do not require convexity and continuously differentiable
properties of objective functions.
Therefore, the KKT condition and genetic algorithm are used to solve the BLPM in this paper
(Figure 3). If the lower level function is convex, the KKT condition can be used to transform the bi-level
programming model into a single programming model. The non-convex situation is not considered in this
paper, but is considered for further research. Based on the operational data and financial data of a public
transport operator, the single level programming model can be solved via a genetic algorithm (GA). GA is
robust and widely used to solve nonlinear programming problems. The process of GA is mainly divided
into four steps: initialization, choice, duplication, and termination judgment. The detailed process can be
found in [3].
Figure 3. Solution flow diagram.
Hypothesizing that the lower function is convex, and other situations will be discussed in further
research. If the involved route is more attractive and the difference between routes is less than the profits
returned to passengers, the lower objective function is convex [3]. Then, the KKT equivalent conditions of
the lower level programming model (Equations (8)–(11)) are presented as follows:
{𝑞𝑖,𝑘≤𝑇𝐶𝑎𝑝𝑖,𝑘𝐹𝑖,𝑘min≤𝐹𝑖,𝑘≤𝑁𝑇𝑖,𝑘2𝑟𝑡𝑖,𝑘𝜆1(𝑞𝑖,𝑘−𝑇𝐶𝑎𝑝𝑖,𝑘)=0𝜆2(𝐹𝑖,𝑘−𝐹𝑖,𝑘min)
(𝐹𝑖,𝑘−𝑁𝑇𝑖,𝑘2𝑟𝑡𝑖,𝑘)=0∇[(𝑄𝑝𝑢𝑏𝑙𝑖𝑐+𝑄𝑐𝑎𝑟)∑𝑖𝐶𝑖⋅𝑒𝜃𝑉𝑘∑𝑘𝑒𝜃𝑉𝑘−𝐶𝑝𝑟𝑖𝑣𝑎𝑡𝑒0(𝑄𝑝𝑟𝑖𝑣𝑎𝑡𝑒)
+∑𝑠∑𝑡𝑄𝑐𝑎𝑟𝑠𝑡𝐶𝑐𝑎𝑟𝑠𝑡]+𝜆1∇(𝑞𝑖,𝑘−𝑇𝐶𝑎𝑝𝑖,𝑘)+𝜆2∇[(𝐹𝑖,𝑘−𝐹𝑖,𝑘min)(𝐹𝑖,𝑘−𝑁𝑇𝑖,𝑘2𝑟𝑡𝑖,𝑘)]
+𝜇1∇(𝑄𝑝𝑢𝑏𝑙𝑖𝑐−∑𝑖,𝑘𝑄𝑖,𝑘*(𝑓𝑖,𝑘,𝐹𝑖,𝑘))=0𝜆1≥0,𝜆2≥0,𝜇1≠0;𝑖=𝑟𝑜𝑢𝑡𝑒𝑛𝑢𝑚𝑏𝑒𝑟;𝑘=𝑏𝑢𝑠,𝑏𝑟𝑡,𝑠𝑢𝑏𝑤𝑎𝑦
(17)

6. Case Study
Jinan city, the capital of Shandong Province, China, was selected for this case study. The urban
population of Jinan was 4.4 million at the end of 2015 [70]. The total amount of motor vehicles of Jinan
had reached 1.67 million by the end of 2015, 1.48 million of which were private cars. The public transport
of Jinan city has developed rapidly in recent years due to its struggle to be the National Public Transit City
of China. There were about 12,000 staff, 5200 buses in operation in the Jinan Public Traffic Company by
the end of 2015. The public transport system had 240 bus routes (eight BRT routes included) with a total
length of 4100 kilometers and the network length of 1280 kilometers, and carried more than 2.4 million
passenger trips daily (the BRT volume was about 0.3 million per day). At the same time, there were about
8000 taxis by the end of 2015 in Jinan city. According to the comprehensive transportation survey of
Jinan city in 2011, the public transit mode share rate (excluding walking) reached 29.69% (Figure 4).
Walking, bikes, and electric bikes are green traffic modes, and these traffic modes should be maintained.
However, the electric bike has caused many safety problems due to its high speed and simple bumper.
Based on the travel survey of Jinan city, travel time and travel cost were similar for the same trip by
private car, business car, or taxi. Therefore, the three travel modes are considered together. Company
buses and school buses are also operated by the public transport company, and they can be considered
together with buses. In the following section, we mainly analyze the competitions among buses, BRT, and
cars. BRT5, bus line 49, and car/taxi routes were chosen in this section to form a network between Jinan
Railway Station and the Yan-Shan Overpass (Figure 5).
Figure 4. Mode share rate of Jinan city based on the survey in 2011 [71]. (a) Walking is included in mode
split; (b) Walking is excluded in mode split.

Figure 5. Distribution of chosen routes and stops. Note: Lr demonstrates the distance between stops of
bus rapid transit (BRT)5; Lb presents the distance between the stops of bus line 49; Lc means the length
of car/taxi route. Lrw is walk distance to BRT5; Lbw is walk distance to Bus 49; O is origin and D is
destination.
Suppose the private sector is involved in the service provision of BRT routes. Parameters in the
BLPM under a lease contract were calculated according to the operational data and financial data of the
Jinan Public Traffic Company and Jinan city travel survey in 2011 [71]. Parameter values are listed
in Table 3.
Table 3. Parameter values of the bi-level programming model.

The GA toolbox of MATLAB (2012a version, MathWorks Inc., Natick, MA, USA) was used for the
model solution under several comparable conditions, such as with and without the private capital
investment in public transport, and whether or not to give some returns to passengers. The minimum
ticket price in Table 3 was treated as the specific cost passengers were willing to pay.
 Model results in three comparable conditions before and after considering car competition with
public transport are listed in Table 4. The last column ‘objective value’ is calculated with MATLAB under
the constraints of ticket fare and departure frequency. The solution algorithm was given in Section 5.
Objective value means the optimal welfare of the public transport system under the given value range of
each parameter. The other columns are the variable values when the objective value reaches an optimal
value. For example, the second row of Table 4 means without car competition and without private capital
investment, the optimal objective value (welfare) of the public transport system is 2737 yuan; at this time,
the bus ticket price should be 1 yuan, the BRT ticket price should be 2 yuan, the bus departure frequency
should be 15 buses/hour, and the BRT departure frequency should be 20 BRT/hour. For the three
investment situations, the objective value, respectively, decreases after the private car competes with
public transport under the same public transport travel demand (the objective value changes from 2737 to
2271 in the situation without private investment, 3457 drops to 2371 under the traditional investment, and
from 3488 to 2396 in the third investment situation).
Table 4. Results of different investment types.

When the transportation system is optimal under each situation, bus and BRT ticket prices all reach
the lowest level at the same time the departure frequency reaches the highest level. The objective value
of private capital investment with returns (considering passenger value) is the greatest one (2396 yuan
when considering car competition, and 3488 yuan without car competition), while the public transport
system has the least welfare (2271 yuan when considering car competition, and 2737 yuan without car
competition) without private capital investment, whether or not considering car competition with public
transport. The results show that private capital investment is beneficial to the total welfare of the public
transport system, and private capital investment in public transport considering passenger value
(returning some profits to passengers) is superior to the traditional investment without considering
passenger value. The system welfare under proposed strategy is 2396 yuan, 125 yuan higher than that
without private investment (2271 yuan). The system welfare has increased by 5.5%. The lease fee for the
private sector should not be higher than 225 yuan/day (the present fee is 100 yuan/day, meaning there is
a growth space of 125 yuan). Otherwise, the system welfare will be less than that without private
investment.
Table 5 shows the results under different passenger demands when other parameters are fixed. If
the passenger flow is less than 200 in this case, it does not satisfy the profit constraint. Therefore, travel
demand should be carefully considered when the private capital invests in public transport. The lease fee
and subsidies should be adjusted reasonably. Obviously, the results also depend on the ticket price and
other costs. After all, the model can show a reasonable tendency. For a given city and fixed transport
system, it can analyze the investment feasibility more precisely than qualitative analysis.
Table 5. Model solution results under different passenger demands.

Benefits returned to passengers will attract more people to take the bus. Since the benefit rate of
cash flow in accounts of bus cards is higher than that of bank interest, there will be additional returns for
passengers. However, in order to satisfy the benefit rate of the private sector, there should be some
critical point. Table 6 shows that the critical point is 0.9 in this case. As the return ratio increases, the
system objective value increases gradually. The system welfare when return rate reaches the critical point
is 2416.9 yuan; the value increases by 1.9% than without returns to passengers. Additionally, the return
critical point is determined by all of the parameters and constraint conditions. This case is only one bus
route and one BRT route; the total welfare of the whole city’s public transport system will be considerable.
Table 6. Model solution results at different return ratios.

The quantitative effects of private capital investment in public transport are obvious. The critical
points of the lease fee, passenger flow, and return rate can also be calculated via the model. The results
are important for decision-makers and private sectors. The BLPM is reasonable for private capital
investment in public transport considering passenger value in terms of the process of the solution and
results analysis. Specific parameters can be regulated and controlled, and policy suggestions can also be
recommended based on the model.
7. Discussion and Conclusions
This study provides a comprehensive view of private capital investment. Private capital investment
in urban public transport is limited in scope [3] to mainly infrastructure construction and operations. Since
investments in construction are well-addressed, investments in operations are particularly discussed in
the paper. Due to unsustainable finance, previous investments in public transport operation have largely
failed [3]. As such, a new investment strategy considering passenger value is proposed in the paper. In
order to quantify the effects of the proposed investment scheme, the bi-level programming quantitative
evaluation model of private capital investment in public transport considering passenger value was
established, a solution algorithm was designed, and the model validity was verified through a case
analysis of Jinan public transport of China. The research results provide a theoretical basis for quantifying
the impacts of private capital investment in public transport and designing more efficient investment
schemes.
Through the results shown in Table 4, Table 5 and Table 6, we can obtain the critical points of the
lease fee, passenger flow, and return rate. The system welfare has increased by 5.5% after considering
passenger value. The variations of objective values are smooth when critical parameters (ticket price,
passenger flow, return rate) take different values. There are no big fluctuations for the objective values.
Therefore, the critical parameters are robust. Additionally, the total welfare of the whole city’s public
transport system will be considerable if considering passenger value. Private capital investment is
beneficial to the total social welfare of the public transport system, and the investment model considering
passenger value is superior to the traditional investment. The new investment scheme is better for
sustainable development of public transport. Private investment considering passenger value should be
encouraged. The proposed investment strategy increased the total social welfare of the public transport
system and increased the total benefits of private sector when considering passenger value. The
proposed investment strategy is beneficial to the financial sustainability of public transport development.
This research is related to the topic of sustainability especially the financially sustainable development of
public transport. In this case, the lease fee for the private sector should not be higher than 225 yuan/day.
Additionally, passenger flow should be higher than 200. The above analyses show that the competition
among different traffic modes, public transport travel demand, government subsidies, and returns to
passengers should be considered carefully. The quantitative results of the case study are valuable for
decision-makers and the private sector. The VFM analysis can also give financial results, but VFM only
considers the financial aspect. For public transport, traveler’s choice behavior is very important, which will
affect the number of passengers [39].
Additionally, Table 4 also shows private cars are bad for public transport system, and travelers
should be encouraged to use public transport through improving the service quality and passenger
returns. Only in this way can the success rate of private sector investment in public transport be improved
efficiently.
Although the case study is simple, the model construction and model solution provide a foundation
for further research and practical application. If the operational data and financial data of public transport
in another city in China and elsewhere are collected, BLPM can be solved easily according to the solution
algorithm in the manuscript. Collecting the related data is not a problem for government officials in the
management department of public transport. The model established in the paper can be transplanted to
other cities if the parameter values in Table 3 are collected. The paper also has some shortcomings. It is
obvious there is competition and cooperation among different public transport modes. Cooperation is
important in the public transport network due to transfers between different public transport modes. The
intermodal issues are very important; however, the issues will make the complexity of BLPM increase
rapidly. The model, at present, only considers competition among different public transport modes,
although in the case study, competition is more common for the same origin and destination. Perhaps
cooperation is more important if there are transfer situations. Xue and Guan [3] also established BLPM for
private investment in public transport in previous research and chose a binary logit model as the travel
mode choice model. They also only considered the competition between buses and BRT. In this study,
private cars and taxis were added to the competition among the traffic model, and a multinomial logit
model was chosen as the traffic mode choice model. Intermodal issues could be considered in further
research in a more complicated network.
Based on the BLPM established in the paper, more detailed and further research can be developed.
In fact, each parameter is not certain as a fixed value in the sample case. This is the uncertainty or
 stochastic bi-level programming problem which will be explored in following studies. The solution problem
of BLPM when the lower function is not convex is also worthy of discussion. Additionally, the strategies
under more complicated networks and intermodal issues need further research and design in detail. After
all, the BLPM constructed in the paper laid the foundation for further research and provides a quantitative
analysis tool which is more reasonable than the value for money analysis.

Acknowledgments
This research was sponsored by the National Natural Science Foundation of China (Grant No.
51338008, 51378036, 51308018 and 51308015). The authors give thanks to the Jinan Public Traffic
Company for operational data. Authors are also very grateful for the comments from the editor and the
two anonymous reviewers.

Author Contributions
The individual contribution and responsibilities of the authors are listed as follows: Hongzhi Guan
designed the research, developed the model; Yunqiang Xue collected the data, analyzed the data,
conducted model validation and wrote the paper; Jonathan Corey revised the manuscript, provided some
comments and helped edit the manuscript; Heng Wei gave some comments and revised the manuscript;
Hai Yan gave some comments and edited the manuscript.

Conflicts of Interest
The authors declare no conflict of interest.

References
1. Malandraki, G.; Papamichail, L.; Papageorgiou, M.; Dinopoulou, V. Simulation and Evaluation
of a Public Transport Priority Methodology. Trans. Res. Procedia 2015, 6, 402–410. [Google
Scholar] [CrossRef]
2. Jara, D.; Gschwender, A.; Ortega, M. The impact of a financial constraint on the spatial
structure of public transport services. Transportation 2014, 41, 21–36. [Google Scholar]
[CrossRef]
3. Xue, Y.; Guan, H.; Corey, J.; Qin, H.; Han, Y.; Ma, J. Bi-level Programming Model of Private
Capital Investment in Urban Public Transportation: Case Study of Jinan City. Math. Probl.
Eng. 2015, 2015, 1–12. [Google Scholar] [CrossRef]
4. Chan, A.; Lam, P.; Chan, D.; Cheung, E.; Ke, Y. Privileges and attractions for private sector
involvement in PPP projects. Chall. Oppor. Solut. Struct. Eng. Constr. 2010, 2010, 751–755.
[Google Scholar]
5. Cox, W. Competitive Participation in U.S. Public Transport: Special Interests versus the Public
Interest. In Proceedings of the 8th International Conference on Competition and Ownership
in Land Passenger Transport, Rio de Janeiro, Brazil, 15–18 September 2003.
6. Daube, D.; Vollrath, S.; Alfen, H. A Comparison of Project Finance and the Forfeiting Model as
financing forms for PPP projects in Germany. Int. J. Proj. Manag. 2008, 26, 376–387. [Google
Scholar] [CrossRef]
7. Costa, A.; Fernandes, R. Urban public transport in Europe: Technology diffusion and market
organisation. Transp. Res. A 2012, 46, 269–284. [Google Scholar] [CrossRef]
8. Matsunaka, R.; Tetsuharu, O.; Nakagawa, D.; Nagao, M.; Nawrocki, J. International
comparison of the relationship between urban structure and the service level of urban public
transportation—A comprehensive analysis in local cities in Japan, France and
Germany. Transp. Policy 2013, 30, 26–39. [Google Scholar] [CrossRef] [Green Version]
9. Medda, F.R.; Carbonaro, G.; Davis, S.L. Public private partnerships in transportation: Some
insights from the European experience. IATSS Res. 2013, 36, 83–87. [Google Scholar]
[CrossRef]
10. Lu, Y. Research on the Reform of Marketization in Urban Public Transport Industry; Economy
& Management Publishing House: Beijing, China, 2014. [Google Scholar]
11. Rossi, M.; Civitillo, R. Public Private Partnerships: A general overview in Italy, in 2nd World
Conference on Business, Economics and Management. WCBEM 2013, 2014, 1–10. [Google
Scholar]
12. Osei-Kyei, R.; Chan, A.P.C. Review of studies on the Critical Success Factors for Public-
Private Partnership (PPP) projects from 1990 to 2013. Int. J. Proj. Manag. 2015, 33, 1335–
1346. [Google Scholar] [CrossRef]
13. English, L.M. Public Private Partnerships in Australia: An Overview of Their Nature, Purpose,
Incidence and Oversight. UMSW Law J. 2006, 29, 250–263. [Google Scholar]
14. World Bank; Asian Development Bank; Inter-American Development Bank. Public-Private
Partnerships Reference Guide—Version 2.0; World Bank Group: Washington, DC, USA, 2014.
[Google Scholar]
15. De Los Rios-Carmenado, I.; Ortuno, M.; Rivera, M. Private-Public Partnership as a Tool to
Promote Entrepreneurship for Sustainable Development: WWP Torrearte
Experience. Sustainability 2016, 8, 199. [Google Scholar] [CrossRef]
16. Bidne, D.; Kirby, A.; Luvela, L.J.; Shattuck, B.; Standley, S.; Welker, S. The Value for Money
Analysis: A Guide for More Effective PSC and PPP Evaluation; American University:
Washington, DC, USA, 2013; pp. 1–60. [Google Scholar]
17. Ho, P.H.K. Development of Public Private Partnerships (PPPs) in China. Surv. Times 2006, 15,
1–5. [Google Scholar]
18. Takim, R.; Ismail, K.; Nawawi, A.; Jaafar, A. The Malaysian Private Finance Initiative and
Value for Money. Asian Soc. Sci. 2009, 5, 103–111. [Google Scholar] [CrossRef]
19. Jung, S.; Huynh, D.; Rowe, P.G. The pattern of foreign property investment in Vietnam: The
apartment market in Ho Chi Minh City. Habitat Int. 2013, 39, 105–113. [Google Scholar]
[CrossRef] [Green Version]
20. Fox, E.M. An Anti-Monopoly Law for China-Scaling the Walls of Protectionist Administrative
Restraints. Antitrust Law J. 2007, 74, 1–31. [Google Scholar]
21. Mu, R.; de Jong, M.; Koppenjan, J. The rise and fall of Public-Private Partnerships in China: A
path-dependent approach. J. Transp. Geogr. 2011, 19, 794–806. [Google Scholar]
[CrossRef]
22. Zhang, S.; Gao, Y.; Feng, Z.; Sun, W. PPP application in infrastructure development in China:
Institutional analysis and implications. Int. J. Proj. Manag. 2015, 33, 497–509. [Google
Scholar] [CrossRef]
23. Chang, Z. Financing new metros-The Beijing metro financing sustainability study. Transp.
Policy 2014, 32, 148–155. [Google Scholar] [CrossRef]
24. Soomro, M.A.; Zhang, X. An Analytical Review on Transportation Public Private Partnerships
Failures. Int. J. Sustain. Constr. Eng. Technol. 2011, 2, 62–80. [Google Scholar]
25. Soomro, M.A.; Zhang, X. Failure Links between Public and Private Sector Partners in
Transportation Public Private Partnerships Failures. J. Traffic Logist. Eng. 2013, 1, 116–121.
[Google Scholar] [CrossRef]
26. Soomro, M.A.; Zhang, X. Evaluation of the Functions of Public Sector Partners in
Transportation Public-Private Partnerships Failures. J. Manag. Eng. 2015, 32, 1–11. [Google
Scholar] [CrossRef]
27. Zhang, X.; Soomro, M.A. Failure Path Analysis with Respect to Private Sector Partners in
Transportation Public-Private Partnerships. J. Manag. Eng. 2015, 32, 1–10. [Google
Scholar] [CrossRef]
28. Winston, C. Government Failure in Urban Transportation. Fiscal Stud. 2000, 21, 403–425.
[Google Scholar] [CrossRef]
29. Murthi, M.; Orszag, J.M.; Orszag, P.R. The Value for Money of Annuities in the UK: Theory,
Experience and Policy; World Bank: Washington DC, USA, 1994. [Google Scholar]
30. Nisar, T.M. Value for money drivers in public private partnership schemes. Int. J. Public Sect.
Manag. 2007, 20, 147–156. [Google Scholar] [CrossRef]
31. Soomro, M.A.; Zhang, X. Value for Money Drivers in Transportation Public Private
Partnerships. In Proceedings of the 24th International Project Management Association
(IPMA) World Congress, Istanbul, Turkey, 1–3 November 2010; pp. 1–6.
32. Berger, P.; Hawkesworth, I. How To Attain Value for Money: Comparing PPP and Traditional
Infrastructure Public Procurement. OECD J. Budg. 2011, 1, 1–56. [Google Scholar]
[CrossRef]
33. European PPP Expertise Centre (EPEC). Value for Money Assessment; European PPP
Expertise Centre: Luxembourg, 2015. [Google Scholar]
34. Mounter, N.; Annema, J.A.; van Wee, B. Managing the insolvable limitations of cost-benefit
analysis: Results of an interview based study. Transportation 2015, 42, 277–302. [Google
Scholar] [CrossRef]
35. Proost, S.; Dunkerley, F.; van der Loo, S.; Adler, N.; Brocker, J.; Korzhenevych, A. Do the
selected Trans European transport investments pass the cost benefit
test? Transportation 2014, 41, 107–132. [Google Scholar] [CrossRef]
36. Macario, R. Future challenges for transport infrastructure pricing in PPP arrangements. Res.
Transp. Econ. 2010, 30, 145–154. [Google Scholar] [CrossRef]
37. Morallos, D.; Amekudzi, A. A Review of Value-for-Money Analysis for Comparing Public-
Private Partnerships with Traditional Procurements. In Proceedings of the Transportation
Research Board 87th Annnual Meeting, Washington, DC, USA, 13–17 January 2008; pp. 1–13.
38. Peng, W. Achieving Value for Money: An Analytic Review of Studies on Public Private
Partnerships. Constr. Res. Congr. 2014, 2014, 1–10. [Google Scholar]
39. Yang, H.; Kong, H.Y.; Meng, Q. Value-of-time distributions and competitive bus
services. Transp. Res. E 2001, 37, 411–424. [Google Scholar] [CrossRef]
40. Pedro, A.L.A. The economic value of bus subsidy. Transp. Res. Procedia 2013, 8, 247–258.
[Google Scholar]
41. Wan, X.; Chen, J.; Wang, W. Travel Characteristics and Mode Choice Behavior of Public
Transport in China. Int. Conf. Transp. Eng. 2009, 2009, 1–6. [Google Scholar]
42. Tang, R.; Sun, H. Discussion on involvement modes and fields of private capital introducted
into transportation industry. Financ. Acc. Commun. 2012, 8, 10–16. [Google Scholar]
43. Cruz, C.O.; Marques, R.C. Infrastructure Public-Private Partnerships: Decision, Management
and Development; Springer: Berlin, Germany, 2013. [Google Scholar]
44. Xu, S. Applying Pubic-Private-Partnership to Chinese Subway Infrastructure; Delft University
of Technology: Delft, The Netherlands, 2008. [Google Scholar]
45. Percoco, M. Quality of institutions and private participation in transport infrastructure
investment: Evidence from developing countries. Transp. Res. A 2014, 70, 50–58. [Google
Scholar] [CrossRef]
46. Seybold, P.B. The Customer Revolution: How to Thrive When Customers Are in Control;
Crown Business: New York, NY, USA, 2001. [Google Scholar]
47. Verhoef, P.C.; Lemon, K.N. Successful customer value management: Key lessons and
emerging trends. Eur. Manag. J. 2013, 31, 1–15. [Google Scholar] [CrossRef] [Green
Version]
48. Kahreh, M.; Tive, M.; Babania, A.; Hesan, M. Analyzing the applications of customer lifetime
value (CLV) based on benefit segmentation for the banking sector. Procedia Soc. Behabioral
Sci. 2014, 109, 590–594. [Google Scholar] [CrossRef]
49. Ulaga, W.; Chacour, S. Measuring customer-perceived value in business markets: A
prerequisite for marketing strategy development and implementation. Ind. Mark.
Mana. 2001, 30, 525–540. [Google Scholar] [CrossRef]
50. Brown, S.A. Customer Relationship Management: A Strategy Imperative in the World of E-
Business; John Wiley & Sons: New York, NY, USA, 2000. [Google Scholar]
51. Gerpott, T.J.; Rams, W.; Schindler, A. Customer retention, loyalty, and satisfaction in the
German mobile cellular telecommunication market. Telecommun. Policy 2001, 25, 249–269.
[Google Scholar] [CrossRef]
52. Chou, J.S.; Yeh, C.P. Influential constructs, mediating effects, and moderating effects on
operations performance of high speed rail from passenger perspective. Transp.
Policy 2013, 30, 207–219. [Google Scholar] [CrossRef]
53. Wong, W.G.; Han, B.M.; Ferreira, L.; Zhu, X.N.; Sun, Q. Evaluation of management strategies
for the operation of high-speed railways in China. Transp. Res. A Policy Pract. 2002, 36, 277–
289. [Google Scholar] [CrossRef] [Green Version]
54. Xue, X. White Book of Jinan Public Traffic Development in Year 2011; Jinan Urban Transport
Research Center: Jinan, China, 2012. [Google Scholar]
55. Wei, M.; Sun, B.; Jin, W. A Bi-level Programming Model for Uncertain Regional Bus Scheduling
Problems. J. Transp. Syst. Eng. Inf. Technol. 2013, 13, 106–113. [Google Scholar]
[CrossRef]
56. Zhao, X.; Yang, J. Research on the Bi-level Programming Model for Ticket fare pricing of
Urban Rail Transit Based on Particle Swarm Optimization Algorithm. In Proceedings of the
13th COTA International Conference of Transportation Professionals, Shenzhen, China, 13–16
August 2013; pp. 633–642.
57. Moshe, B.-A.; Bierlaire, M. Discrete Choice Analysis: The Theory and Application on Travel
Demand; MIT Press: Cambridge, MA, USA, 2010. [Google Scholar]
58. Train, K.E. Discrete Choice Methods with Simulation, 2nd ed.; Cambridge University Press:
Cambridge, UK, 2009. [Google Scholar]
59. Li, J.; Lai, X.; Yu, Z. A paired combinatorial logit route choice model with probit-based
equivalent i mpedance. J. Transp. Syst. Eng. Inf. Technol. 2013, 13, 100–105. [Google
Scholar]
60. Zhang, G. Enhanced traffic information dissemination to facilitate toll road utilizaiton: A
nested logit model of a stated preference survey in Texas. Transportation 2014, 41, 231–
249. [Google Scholar] [CrossRef]
61. Paulssen, M. Values, attitudes and travel behavior: A hierarchical latent variable mixed logit
model of travel mode choice. Transportation 2014, 41, 873–888. [Google Scholar]
[CrossRef]
62. Colson, B. Mathematical Programs with Equilibrium Constraints and Nonlinear Bi-Level
Programming Problems; University of Namur: Namur, Belgium, 1999. [Google Scholar]
63. Colson, B.; Marcotte, P.; Savard, G. A trust-region method for nonlinear programming:
Algorithm and computational experience. Cumput. Optim. Appl. 2005, 30, 211–227.
[Google Scholar] [CrossRef]
64. Colson, B.; Marcotte, P.; Savard, G. An overview of bilevel optimization. Ann. Oper.
Res. 2007, 153, 235–256. [Google Scholar] [CrossRef]
65. Yokota, T.; Gen, M.; Li, Y.; Kim, C.E. A genetic algorithm for interval nonlinear integer
programming problem. Comput. Ind. Eng. 1996, 31, 913–917. [Google Scholar]
[CrossRef]
66. Demps, S. Foundation of Bi-Level Programming; Kluwer Academic Publishers: Dordrecht, The
Netherlands, 2003. [Google Scholar]
67. Baky, I.A. Solving multi-level multi-objective linear programming problems through fuzzy
goal programming approach. Appl. Math. Model. 2010, 34, 2377–2387. [Google Scholar]
[CrossRef]
68. Tang, J.; Zhang, G.; Wang, Y.; Wang, H.; Liu, F. A hybrid approach to integrate fuzzy C-means
based imputation method with genetic algorithm for missing traffic volume data
estimation. Transp. Res. C 2015, 51, 29–40. [Google Scholar] [CrossRef]
69. Yang, H.; Bell, M.G. Transport bi-level programming problems: Recent methodological
advances. Transp. Res. B Methodol. 2001, 35, 1–4. [Google Scholar] [CrossRef]
70. Jinan, B.O.S. Jinan Statistical Year Book 2015; China Statistic Press: Beijing, China, 2016.
[Google Scholar]
71. Xue, Y.; Guan, H.; Qin, H.; Liu, T.; Gong, L. The study on key factors influencing public
transport share rate based on disaggregate model-Jinan as an example. In Proceedings of
the 14th COTA International Conference of Transportation Professionals, Changsha, China,
4–7 July 2014; pp. 1–13.
© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution (CC
BY) license ( http://creativecommons.org/licenses/by/4.0/).