Professional Documents
Culture Documents
Decision Support
a r t i c l e i n f o a b s t r a c t
Article history: In this paper, we consider revenue maximization problems for service systems with heterogeneous cus-
Received 5 March 2018 tomers. To extract meaningful insights, we consider a simplified two queue system where each server
Accepted 25 April 2019
charges an admission price to its customers. The customers differ in their cost for unit delay and this
Available online 3 May 2019
customer heterogeneity is modeled as a random variable with distribution F. On arrival, the customers
Keywords: must choose a server that offers the lowest sum of admission price and the expected delay cost.
Pricing Given the admission prices at the two servers and the explicit knowledge of the distribution F, we
Revenue maximization first characterize the Wardrop equilibrium routing for the heterogeneous customers. We then consider a
Wardrop equilibrium monopoly system where both the servers belong to a single operator. For this system we characterize the
Duopoly revenue maximizing program for the monopoly operator and outline its properties. We then consider the
Queueing system duopoly problem where each server competes with the other server to maximize its revenue rate. For
the special case when the servers have identical service capacity, we obtain the necessary condition for
existence of symmetric Nash equilibrium prices.
For the analysis of the monopoly and the duopoly problem, the heterogeneity distribution F is as-
sumed to be explicitly known. However, for most practical scenarios, the functional form of F may not be
known to the system operator and in such cases, the revenue maximizing prices cannot be determined. In
the last part of the paper, we provide an elementary method to estimate the distribution F. This method
relies on suitably varying the admission prices based on the properties of the Wardrop equilibrium. We
also illustrate this method with some suitable numerical examples.
© 2019 Elsevier B.V. All rights reserved.
1. Introduction Many service systems have emerged that exploit the heteroge-
neous nature of customer preferences and use it to their advan-
In many service systems, the quality of service received is char- tage. As an example, airlines have a separate priority queue which
acterized by the queueing delay that is experienced by the cus- is available to delay sensitive customers after payment of an ad-
tomers in the system. Examples of such service systems include ditional fee. Some airports even offer priority security checks for
road and transport systems, health-care systems, computer sys- passengers willing to pay an additional fee. Passport or visa issu-
tems, call centers and communications systems. The customers of ing agencies all over the world offer the facility of quick processing
such systems are usually sensitive to the service quality or the de- of their passport or visa with a payment of an additional fee. Sim-
lay experienced in these systems. Further they have non-identical ilarly, amusement parks such as Disneyland or Universal Studios
preferences to the delay experienced. It is often beneficial for offer a priority queue for customers willing to pay a higher ad-
the service system to account for these heterogeneous customer mission price. In service systems with such a price priority struc-
preferences in any optimization concerning the use of system re- ture, the users of the system are faced with a conflicting choice
sources. In this paper, we consider the problem of exploiting this of whether to pay the additional fee or not. Typically one could
heterogeneous nature of customer preferences for revenue maxi- expect that customers with higher sensitivity or cost for unit de-
mization in parallel server systems. lay would have a higher willingness to pay the extra fee. However,
this choice for a customer is also influenced by the congestion level
∗ in the priority queue which in turn depends on the choice of ser-
Corresponding author.
E-mail addresses: tejaspbodas@gmail.com (T. Bodas), dmanju@ee.iitb.ac.in vice by the other customers based on their cost for unit delay thus
(D. Manjunath). complicating the matters.
https://doi.org/10.1016/j.ejor.2019.04.041
0377-2217/© 2019 Elsevier B.V. All rights reserved.
T. Bodas and D. Manjunath / European Journal of Operational Research 278 (2019) 686–698 687
Similarly, the firm or the service system offering a separate pri- identify the necessary condition for existence of symmetric Nash
ority queue is faced with the problem of determining the admis- Equilibrium prices at the two servers.
sion fee for this queue. Very often (as in case of amusement parks For both the revenue maximization problems, a crucial assump-
or airline security queues), the objective of the firm is to maxi- tion is that the optimizer of the system has the complete knowl-
mize the revenue generated. Again, the revenue depends on the edge of the delay cost distribution for the arriving customers. How-
rate of users choosing to make this extra payment which again de- ever, in most practical scenarios, the distribution F(·) characterizing
pends on the users cost for unit delay. A similar question arises the delay cost for a customer may not be completely known to the
in case of two competing firms (such as two competing amuse- system. While the revenue maximizing strategy explicitly depends
ment parks) with congestion sensitive customers. A higher admis- on F(·), lack of any knowledge of F(·) makes it difficult to ascer-
sion price makes a firm more attractive to the congestion sensi- tain a revenue optimal admission price. As a simple scenario, con-
tive customer but risks loosing revenue to the other firm in case sider an amusement park where the arriving customers are of two
most of the customers are inherently less sensitive to congestion. types, customers with a high delay cost value and those with a
In this paper, we are concerned with the impact of such cus- low delay cost per unit time. Now suppose that the precise values
tomer heterogeneity not only on the equilibrium choice of queue of the delay costs and the proportion of users of each type are not
for the customers but also on the impact of this heterogenous know. In such a scenario, lack of essential information on the delay
delay cost for customers on the revenue maximizing prices for cost structure would severely affect the ability of the amusement
monopolistic (airport security or passport issuer) and competing park to optimize its revenue. In scenarios such as the one outlined
firms. above, and in the absence of any sophisticated delay cost revealing
We answer a variety of questions like the ones discussed above, mechanism, it is imperative for the service system to in some way
by considering a simplified two server system where customers estimate the unknown distribution F based on the changes in the
have heterogeneous cost for unit delay. We assume that each equilibrium arrival rates which is in turn based on the changes in
server has an associated queue for the customers to wait and that the admission price. In the last section of this paper, we provide
the scheduling discipline at each server is work conserving and an elementary method to estimate F(·) by varying the admission
does not discriminate between customers on the basis of their prices and observing the change in the equilibrium traffic routing.
preference for delay. The servers charge an admission price to ev- We make use of the properties of the Wardrop equilibrium to iden-
ery customer joining its queue. We assume that the queues are tify a procedure to extract missing information on the distribution
not observable and only the expected delay as a function of the F. This information may prove to be beneficial for service systems
arrival rate is available. Further the expected delay at any server is for the underlying revenue maximization.
monotone increasing in the arrival rate of customers to that server.
The customers that use the system are strategic and make an in- 1.1. Related work
dividually optimal queue-joining decision. The customers differ in
their cost for unit delay which is characterized by a random vari- Classical monopoly models have been well studied for the case
able with distribution denoted by F. The cost of service at a server of single server queues with homogeneous customers. One of the
is the sum of the admission price and the delay cost at that server. first work to analyze such a model is Naor (1969). This model con-
To keep the discussion as simple as possible, we additionally as- siders a single server queueing system where homogeneous cus-
sume that customers do not balk from the system without obtain- tomers obtain a reward after service completion. The queue is
ing service. observable to arriving customers who choose to either join the
As our first revenue maximization problem, we consider a queue or balk. For such a system, the revenue maximizing ad-
monopoly and assume that the two servers belong to the same mission price was first obtained in Naor (1969). Subsequent works
operator. The objective here is to maximize the total revenue rate, have analyzed the revenue maximization problem for various mod-
i.e., the sum of the revenue rate from the two servers. As outlined els such as a multiserver queue (Edelson & Hilderbrand, 1972),
earlier, our formulation can be used to analyze monopoly systems GI/M/1 queue (Yechiali, 1971), customers with heterogeneous ser-
such as government agencies issuing documents (passport, visas, vice valuations (Larsen, 1998) and queue length dependent prices
driving license), amusement parks (in small cities that typically (Chen & Frank, 2001). While the above models assume that the
have no competitors), airport security queues etc. It is not diffi- queue lengths are observable, (Edelson & Hilderbrand, 1975) were
cult to see that a revenue maximizing strategy for the monopoly the first to consider the revenue maximization problem for the
is to keep both the admission prices as high as possible. This is case when queues are not observable. See (Bradford, 1996; Chen &
because customers cannot balk and are hence required to choose Frank, 2004; Masuda & Whang, 1999; Mendelson, 1985; Mendel-
one of the servers for service. For a meaningful formulation for a son & Whang, 1990) for some other single server revenue max-
system where customers cannot balk, we assume that the admis- imization models. The key difference between the above litera-
sion price at one of the servers is a priori fixed to a reasonable ture and our model is that we assume customers to have hetero-
value (for example, in the case of the airport security queue, this geneous preference for the delay experienced. This feature makes
is zero). Further, we assume that each server has enough capac- our model more meaningful but at the same time very difficult to
ity to handle the entire traffic rate without resulting in a system analyze.
breakdown. Our interest is to characterize the revenue maximizing The earliest work analyzing the duopoly model with heteroge-
admission price for different examples of the delay functions at the neous customers was by Luski (1976) and Levhari and Luski (1978).
queue and for different examples of the delay cost distribution F. Luski (1976) is interested in knowing whether the revenue maxi-
As our second revenue maximization scenario, we assume a mizing prices set by the two service systems can be equal. It is
duopoly where each server belongs to separate operators and now observed that when the parameters of the model are such that
has the objective of maximizing its individual revenue rate. While the customers have no incentive to balk, the revenue maximizing
the no balking assumption simplifies the analysis to some extent prices set by two identical servers are equal. This is however not
(without losing on any qualitative properties of the problem), we the case when some of the customers prefer to balk. In this case,
observe that the heterogeneity of the customers makes the cal- the equilibrium revenue maximizing prices are not equal. Levhari
culation of the asymmetric price equilibrium difficult. As part of and Luski (1978) provide a numerical analysis for the problem in-
our main result, we focus on the duopoly problem with identi- troduced in Luski (1976). Armony and Haviv (2003) analyze this
cal servers, and using the properties of the Wardrop equilibrium, problem for the case when the customers are from a finite number
688 T. Bodas and D. Manjunath / European Journal of Operational Research 278 (2019) 686–698
of classes and each class has a distinct cost for unit delay. A nu- 2. Preliminaries
merical analysis of the Nash equilibrium admission prices between
the two competing servers is provided. Chen and Wan (2003) con- We first introduce the notation to be used throughout. Let cj
sider the revenue maximization in a duopoly with a single cus- denote the admission price at Server j where j = 1, 2. The cus-
tomer class. The service system is modeled by M/M/1 queues and tomers arrive according to a homogeneous Poisson process with
the customers are allowed to balk from the system. These assump- rate λ. Let Dj (γ j ) denote the delay function associated with queue j
tions on the system model allows them to obtain the sufficient when the queue arrival rate is γ j , where j = 1, 2. Note that γ1 +
conditions for the existence of Nash equilibrium. Similar condi- γ2 = λ. We assume that Dj is monotone increasing and continu-
tions were found in Dube and Jain (2008) who consider an N- ously differentiable in the interior of its domain with a strictly pos-
player oligopoly with multiclass customers. The customer classes itive derivative. Additionally we assume that the cost function at
differ only in their arrival rates and have the same delay cost per the two server satisfies the inequalities D3− j (0 ) < D j (λ ) < ∞ for
unit time. A differentiated service model is considered by Dube j = 1, 2.
and Jain (2010) where each player now operates two types of ser- We associate with each arriving customer a continuous ran-
vices and each service is used by a dedicated class of customers. dom variable β that quantifies a customer’s sensitivity to delay
Again, the key result in Dube and Jain (2010) is to obtain suffi- or congestion. We shall assume that the delay sensitivity β for a
cient condition for Nash equilibrium prices. Mandjes and Timmer customer is a realization of the random variable β. The customer
(2007) consider a duopoly model with two customer classes dif- arrivals constitute a marked Poisson process of intensity λ × F on
fering in their delay cost. The utility of a queue is a decreasing R × R+ . We assume that F is an absolutely continuous cumulative
function of the number of customers using this server. Given the distribution supported on the interval [a, b] of positive reals. We
admission prices, they provide an algorithm that determines the additionally assume that F(·) is strictly increasing and hence the
equilibrium number of customers of each class at the two servers. density f(·) satisfies f(x) = 0 for x ∈ [a, b].
While the existence and uniqueness of such a customer equilib- A customer with delay cost β entering the system must choose
rium is provided, the existence of Nash equilibrium prices is only a queue j so as to minimize c j + β D j (γ j ). Here γ j is determined
conjectured. Ayesta, Anselmi, and Wierman (2011) consider the through the strategies of all customers. We assume that the quan-
oligopoly pricing game for a single customer class and obtain the tities γ 1 , γ 2 , Dj (·), F(·) and cj , for j = 1, 2 are part of common
necessary and sufficient conditions on the Nash equilibrium prices knowledge. We also assume that the customers do not have ac-
when the queues have identical delay functions. A best-response cess to current or past queue occupancies, or the history of arrival
algorithm is then provided to numerically obtain Nash equilibrium times. The strategy of a customer is restricted to choosing a server
prices. Fan, Kumar, and Whinston (2009) consider a price compe- according to a fixed probability distribution and such joint strate-
tition model between service providers offering shrink-wrap soft- gies are represented by a stochastic kernel, denoted by KW . We in-
ware (SWS) and software as a service (SaaS). They assume a model terpret KW (β , i) as the probability that a customer with delay sen-
where the customers have heterogeneous sensitivity to the imple- sitivity β chooses queue i at equilibrium. For the two server sys-
mentation cost of the software but have identical valuations for tem, the equilibrium kernel KW must satisfy the following Wardrop
the two software services. Additionally, for the SaaS provider, a equilibrium conditions.
cost for service quality is assumed and a choice of the service
quality rate is part of the strategy (along with the price) for the K W (β , i ) ≥ 0 implies ci + β Di (γi ) ≤ c3−i + β D3−i (γ3−i ). (1)
SaaS provider. As part of their main result, the Nash equilibrium In words, this means that if customers with delay cost β choose
prices for the two providers is characterized for a very general set- Server i at equilibrium, then the expected cost for this customer
ting of the parameters. The primary difference between this model at Server i must be at most the expected cost at Server 3 − i for
and that of our paper is that the above model does not involve i = 1, 2. For a kernel KW , note that the arrival rate of customers to
any negative congestion effects on customers choosing the service Server j is given by
providers. It is the absence of a congestion function that allows
b
for explicit expressions for the Nash equilibrium prices. See the
books (Hassin, 2016; Hassin & Haviv, 2003) for a comprehensive
γj = λ K W ( β , j )d F ( β ).
β =a
study of revenue maximization problems for queueing systems un-
der Wardrop equilibrium. Also see (Afeche, 2013; Katta & Sethur- We now provide the following theorem that characterizes the
man, 2005; Van Mieghem, 1995) for other related work analyzing Wardrop equilibrium kernel for the system. (Proof in Appendix)
the impact on service systems with heterogeneous customers dif-
Theorem 1. Define δ i as the probability distribution that puts unit
fering in their delay costs.
mass on i and suppose that the kernel KW satisfies the Wardrop equi-
Most of the monopoly and duopoly models described above,
librium condition of Eq. (1). Then there exists a threshold β 1 with
make simplifying assumptions on the customer classes to charac-
terize the underlying Wardrop equilibrium (Wardrop, 1952). Addi-
β 1 ∈ [a, b] such that
tional simplification of the analysis is obtained by considering con- • when c1 > c2 (resp. c1 < c2 ), we have
vex or M/M/1 type delay functions at the queues. We do not make
such convexity assumptions in this paper. We utilize the structure δ1 (resp. δ2 ) for β ∈ (β1 , b],
K W (β , · ) = (2)
of the Wardrop equilibrium for heterogeneous customers that was δ2 (resp. δ1 ) for β ∈ [a, β1 ].
characterized in Bodas, Ganesh, and Manjunath (2014) to analyze
the two problems. This structure on the equilibrium allows us to Further if β 1 ∈ (a, b) then,
provide an equivalent revenue maximization formulation for both
the monopoly and the duopoly that is simpler to analyze.
c1 + β1 D1 (γ1 ) = c2 + β1 D2 (γ2 ). (3)
The rest of the paper is organized as follows. In the next sec- • When c1 = c2 , is not unique and any kernel K with γ1 =
KW
tion, we shall formalize the notations and provide some prelimi- γ + is a valid Wardrop equilibrium kernel KW where γ + :=
naries. We then formulate the revenue maximization problems in {γ1 : D1 (γ1 ) = D2 (γ2 )}.
Section 3. In Section 4, we consider the monopoly problem and fi-
nally the duopoly problem in Section 5. Finally in Section 6, we Refer Figs. 1 and 2 for a representation of the Wardrop equilib-
illustrate a mechanism based on admission pricing to estimate F. rium kernel for the case when c1 > c2 and c1 < c2 respectively. Here
T. Bodas and D. Manjunath / European Journal of Operational Research 278 (2019) 686–698 689
maximizes its revenue rate Rj . For this duopoly, the revenue opti- γ 1 . Note that g1 (γ 1 ) characterizes the price difference as a func-
mization problem for Server j is as follows. tion of γ 1 . We characterize g1 (γ 1 ) (in Lemma 3) by first character-
izing β 1 (γ 1 ) using Lemmas 1 and 2. For a fixed c2 , we then obtain
max R j (c j , γ j ) = c j γ j (c j , c j − )
cj γ 1 (c2 ) in Lemma 5 that determines the domain of the optimization
subject to 0 ≤ cj ≤ cj (P3 ) problem. To prove this lemma we need to characterize the unique-
given c j− ness of kernel KW for a fixed price difference (Lemma 4). Finally
we characterize our main monopoly program P5 which only de-
where c j− represents the admission price at the server other than pends on the price difference function g1 (·) and not on the exact
j, i.e., c1− = c2 and c2− = c1 . For the duopoly market, the aim is to prices at the individual queues.
obtain the Nash equilibrium set of admission prices to be charged Recall that we make minimal assumptions on the distribution
at the two servers. We shall denote the Nash equilibrium prices F(·) and on the delay cost function Dj (·). For our numerical exam-
by the tuple (c1∗ , c2∗ ). Using the notion of the best response func- ples and also to illustrate the properties of the functions β 1 (·), g1 (·)
tion (see Osborne, 2003 for details), (c1∗ , c2∗ ) can be characterized and c1 (·), we consider the following examples for F(·) and Dj (·). The
as follows. Let Bi (ci− ) denote the admission price at Server i that distribution F(·) considered for our numerical examples are
maximizes the server revenue Ri for a given value of ci− for i = 1, 2.
Clearly, Bi (ci− ) is the maximizer in program P3 and it is easy to see • Uniform distribution over the range [a, b].
that • Exponential distribution with mean τ .
Gamma distribution with shape k and scale θ .
B1 (c2 ) := c1 ≥ 0 : c1 γ1 (c1 , c2 ) ≥ c1 γ1 (c1 , c2 )∀c1 ≥ 0 •
B2 (c1 ) := c2 ≥ 0 : c2 γ2 (c1 , c2 ) ≥ c2 γ2 (c1 , c2 )∀c2 ≥ 0 The delay functions used for numerical examples are
γ
(c1∗ , c2∗ ) = {(c1 , c2 ) : B1 (c2 ) = c1 , B2 (c1 ) = c2 }. • D j (γ j ) = μj . This corresponds to linear delay.
j
However as argued earlier, the closed form expression for • D j (γ j ) = μ 1−γ and μj > λ. This corresponds to M/M/1 type
j j
γ j (c j , c j− ) is not easy to obtain. This makes it difficult to solve delay cost function.
program P3 and obtain the best responses Bi (ci− ) for i = 1, 2. As
a result, obtaining (c1∗ , c2∗ ) is in general not easy. As in the case of The distribution and the delay cost functions outlined above are
the monopoly program, to obtain (c1∗ , c2∗ ), we need to first refor- commonly used to model heterogeneous customers and congestion
mulate program P3 by letting γ j denote the optimizing variable. costs (Refer Bodas et al., 2014; Levhari & Luski, 1978; Luski, 1976).
The corresponding optimization problem is as follows: These examples are not exhaustive and one can assume more com-
plex examples for F and D(·) as long as F is absolutely continuous
max R j (c j (γ j ), γ j ) := c j (γ j )γ j
γj and D(·) is monotone increasing in the arrival rate. We now begin
subject to 0 ≤ γ j ≤ γ j (c j − ) ≤ λ (P4 ) with the following lemma that identifies necessary and sufficient
given c j− . condition on γ 1 when either c1 ≥ c2 or c1 < c2 .
cj (γ j ) can be interpreted as the admission price at Server j that Lemma 1. γ1 ∈ [0, γ + ] iff c1 ≥ c2 while γ1 ∈ (γ + , λ] iff c1 < c2 .
leads to the equilibrium arrival rate of γ j when the other server
charges c j− . Note again that cj (γ j ) will be a function of c j− but we Proof. See Appendix for proof and refer Figs. 1 and 2 for an illus-
do not make this explicit in the notation. Now let γ1∗ (c2 ) denote tration of the lemma.
the maximizer in program P4 for a given value of c2 . Then the best
response c1 is in fact given by the function c1 (γ1∗ (c2 )). Therefore, Next, we express the threshold β 1 of Theorem 1 as a function
once the function c1 (γ 1 ) is characterized, the best response now of γ 1 . Let β 1 (γ 1 ) denote the value of the threshold β 1 (character-
denoted by Bˆ1 (c2 ) satisfies Bˆ1 (c2 ) = c1 (γ1∗ (c2 )). We now have izing KW ) for a given γ 1 such that γ1 = γ + . We have the following
lemma.
(c1∗ , c2∗ ) = (c1 , c2 ) : Bˆ1 (c2 ) = c1 , Bˆ2 (c1 ) = c2
Lemma 2.
where Bˆi (ci− ) = ci (γi∗ (ci− )) and as stated earlier, γi∗ (ci− ) is the
maximizer in program P4 for i = 1, 2. It is therefore clear that F −1 λ−λγ1 for 0 ≤ γ1 < γ + ,
β1 (γ1 ) = (6)
(c1∗ , c2∗ ) can be obtained once we have characterized c1 (γ 1 ). We F −1 γλ1 for γ + < γ1 ≤ λ.
shall analyze the program P4 in detail in Section 5 and explicitly
characterize the functions cj (γ j ) for j = 1, 2 to be able to obtain where F −1 represents the quantile function or the inverse function of
(c1∗ , c2∗ ). the distribution F.
In this section, we will analyze the monopoly program P2. To Note that β 1 (γ 1 ) is not defined in Lemma 2 when γ1 = γ + .
be able to solve program P2, we need to characterize g1 (γ 1 ). This This is because the Wardrop kernel KW with γ1 = γ + is not unique
procedure is outlined below. From Eq. (3) of Theorem 1, we know and hence need not be characterized by a threshold. To avoid
that when β 1 ∈ (a, b), (and hence γ 1 ∈ (0, λ)) we have complications arising due to possibly multiple Wardrop equilibria,
we shall restrict to a threshold type Wardrop equilibrium when
= β1 (D2 (γ2 ) − D1 (γ1 ) ). γ1 = γ + (and hence c1 = c2 ). Hence, for c1 = c2 we assume
Expressing the right hand side of the above equation as a function
δ1 for β ∈ (β1 , b],
of γ 1 and noting that g1 (γ1 ) = c1 (γ1 ) − c2 , we have K (β , 1 ) =
W
(7)
δ2 for β ∈ [a, β1 ].
g1 (γ1 ) = β1 (γ1 )(D2 (λ − γ1 ) − D1 (γ1 ) ) (5)
λ−γ +
where β 1 (γ 1 ) represents the threshold β 1 for a kernel that sat- KW We can now define β1 (γ + ) = F −1 λ and the modified
isfies Theorem 1 and corresponds to an equilibrium arrival rate of β 1 (γ 1 ) is now as follows.
T. Bodas and D. Manjunath / European Journal of Operational Research 278 (2019) 686–698 691
Fig. 3. Illustrating β 1 (γ 1 ) when the servers are not identical. Fig. 5. Illustrating g1 (γ 1 ) when the servers are not identical.
Having obtained β 1 (γ 1 ), we shall now analyze g1 (γ 1 ) that was Remark 3. From Lemma 5, when c2 < −g1 (λ ), we have γ 1 (c2 ) =
defined in Eq. (5). We have the following lemma. γ such that g1 (γ ) = −c2 . From Lemma 3, we know that g1 (γ ) < 0
for γ > γ + . Hence when c2 ≥ 0, we have γ 1 (c2 ) ≥ γ + with strict
Lemma 3. For 0 ≤ γ 1 ≤ λ, g1 (γ 1 ) is continuous and monotonic de- equality when c2 = 0.
creasing in γ 1 . Further, g1 (γ + ) = 0.
Now recall the revenue maximization problem P2. Define γ1∗ as
Proof. See Appendix for proof.
the optimizer for this program with the revenue maximizing ad-
See Fig. 5 for a numerical evaluation of g1 (γ 1 ) when F(·) is ex- mission price at server 1 given by c1 (γ1∗ ). Since RT (c1 (γ1 ), γ1 ) =
ponential with τ = 20 and when the servers have a linear delay c2 λ + (g1 (γ1 )γ1 , γ1∗ must be such that g1 (γ1∗ ) > 0. Now note
with μ1 = 3.3 and μ2 = 4. from Lemma 5 and the subsequent remark that when c2 ≥ 0, we
As stated earlier, the domain of the optimization program P2 have γ 1 (c2 ) ≥ γ + and from Lemma 3 we have g1 (γ 1 ) > 0 for
is determined by γ 1 (c2 ) which is characterized in Lemma 5. Be- γ1 ∈ (0, γ + ). These two facts together imply that γ1∗ ∈ (0, γ + ).
692 T. Bodas and D. Manjunath / European Journal of Operational Research 278 (2019) 686–698
γ
Fig. 6. RT as a function of γ 1 when D j (γ j ) = μjj Fig. 7. RT as a function of γ 1 when D j (γ j ) = μ j 1−γ j
The term c2 λ in RT (c1 (γ 1 ), γ 1 ) is a constant and hence we have We conclude the analysis of the revenue maximization prob-
the following equivalent program for the revenue maximization lem with the following observations made from the two examples
problem. given above.
– We see that for the given examples of F, RT (c1 (γ 1 ), γ 1 ) is a
max g1 (γ1 )γ1 unimodal function in γ 1 . For the three distribution functions, it
γ1 (P5 )
subject to 0 ≤ γ1 ≤ γ + can be shown that F −1 (· ) is differentiable in its arguments. For
such distribution functions with differentiable F −1 (· ), this implies
where g1 (γ 1 ) is given by Eq. (5). that g1 (γ 1 ) and hence RT (c1 (γ 1 ), γ 1 ) is differentiable in γ 1 when
Note from Lemma 3 that g1 (·) is a continuous function of its 0 < γ1 < γ + . From Rolle’s Theorem (Theorem 10.2.7 Tao, 2006),
domain. Program P5 involves maximizing a continuous function this implies that there exists a γ1 ∈ (0, γ + ) such that dγT = 0. A
dR
over a compact set and hence a maximizer γ1∗ exists. The origi- 1
γ1∗ satisfying this equation is the revenue maximizing arrival rate
nal monopoly program P1 has been significantly simplified to the
to server 1. The admission price corresponding to this γ1∗ can now
equivalent program P5. Since g1 (γ 1 ) is strictly decreasing (and
be obtained using Eq. (5).
hence quasi-convex), g1 (γ 1 )γ 1 is in fact a product of two quasi-
– For each of the three distributions, note that we have E β =
convex functions. (However the product of quasi-convex functions
4. However, the Revenue rate RT as a function of γ 1 is distinct in
need not be a quasi-convex function). One can now use standard
all the three cases. This implies that the revenue rate RT depends
non-linear optimization techniques to obtain γ1∗ . To further under-
on the higher moments of the distribution F and not just on its
stand Program P5, we perform a numerical evaluation of g1 (γ 1 )γ 1
mean value.
under a combination of examples for the distribution functions F
– Finally, note that RT depends on admission price through the
and the delay functions Dj (γ j ). We reiterate that our assumptions
addition factor of c2 λ. For different values of c2 , the corresponding
allow us to consider even more complicated examples for F and
γ1∗ does not change. However it is easy to see from Eq. (5) that
D(·). However for convenience, we restrict to the examples that
c1 (γ 1 ) increases linearly in c2 .
were outlined earlier.
γ
Example 1. In this example we shall assume that the D j (γ j ) = μj 5. Duopoly
j
for j = 1, 2. We assume that μ1 = 3.3 and μ2 = 4. Further, the ar-
rival rate λ = 3 and we consider the following three examples for In this section, we shall consider program P4 for revenue max-
the distribution F(·). (1) F has a uniform distribution with support imization in the duopoly system. Much of the analysis in this sec-
on [2,6]. (2) F has an exponential distribution with mean τ = 4 tion follows from that of the previous section. Let γ j , j = 1, 2 de-
and (3) F has a Gamma distribution with the scale k and shape note the optimization variable and represent the admission prices
θ parameters 2 and 2 respectively. Note that β with these three at the respective servers as a function of the arrival rates. Towards
distributions have the same mean. We plot RT (c1 (γ1 ), γ1 ) = c2 λ + this, we continue with the use of the notation cj (γ j ) for j = 1, 2.
g1 (γ1 )γ1 as a function of γ 1 in Fig. 6 where we assume c2 = 1. Note that while in the monopoly case, the admission price c2 was
When F has the uniform distribution, γ1∗ = 0.62. The optimal rev- considered fixed, in the duopoly of this section, it is the strategy
enue rate RT (γ1∗ ) = 4.306 while the admission price c1 (γ1∗ ) max- for the second server and hence will not be a constant. The rev-
imizing RT is 3.106. The corresponding values for the exponential enue function for Server j is given by
distribution are γ1∗ = 0.44, RT (γ1∗ ) = 4.712 and c1 (γ1∗ ) = 4.89 while
the values for the gamma distribution are γ1∗ = 0.51, RT (γ1∗ ) = R j ( c j ( γ j ), γ j ) = c j ( γ j )γ j
4.532 and c1 (γ1∗ ) = 4.
where cj (γ j ) represents the admission price at Server j resulting in
Example 2. In this example, we assume that D j (γ j ) = μ 1−γ where an equilibrium arrival rate of γ j . As noted in the previous section,
j j
cj (γ j ) is a function of c j− , the admission price at the other server.
again μ1 = 3.3 and μ2 = 4. Note that λ < μj for j = 1, 2. The
For a fixed strategy c2 at Server 2, the revenue function R1 (c1 (γ 1 ),
choice of F(·) is as in the previous example. A plot of RT (c1 (γ 1 ), γ 1 )
γ 1 ) can be redefined as
as a function of γ 1 is provided in Fig. 7. When F is uniformly dis-
tributed, γ1∗ = 0.48. The optimal revenue rate RT (γ1∗ ) = 3.83 while R1 (c1 (γ1 ), γ1 ) = (g1 (γ1 ) + c2 )γ1 . (11)
the admission price c1 (γ1∗ ) maximizing RT is 2.72. The correspond-
ing values for the exponential distribution are γ1∗ = 0.33, RT (γ1∗ ) = It can be argued as in the previous section that for a fixed c1
4.21 and c1 (γ1∗ ) = 4.67 while the values for the gamma distribu-
tion are γ1∗ = 0.38, RT (γ1∗ ) = 4.04 and c1 (γ1∗ ) = 3.74. R2 (c2 (γ2 ), γ2 ) = (g2 (γ2 ) + c1 )γ2 (12)
T. Bodas and D. Manjunath / European Journal of Operational Research 278 (2019) 686–698 693
where analyze the Nash equilibrium under the restriction that the two
servers are identical i.e., the average delay at any queue is the
g2 (γ2 ) = β1 (λ − γ2 )(D1 (λ − γ2 ) − D2 (γ2 ) ) (13)
same for the same arrival rate. Under this setting, our interest is
and from Eq. (8), β1 (λ − γ2 ) is as follows: to characterize the symmetric Nash equilibrium such that c1∗ = c2∗ .
F −1 γλ2 for λ − γ + ≤ γ2 ≤ λ, 5.1. Characterizing a symmetric Nash equilibrium
β1 (λ − γ2 ) =
−1 λ−γ2
(14)
F λ for 0 < γ2 < λ − γ + .
In this section, we shall characterize the necessary conditions
It is easy to see that g2 (γ 2 ) is also continuous and strictly de- for the existence of a symmetric Nash equilibrium, i.e., (c1∗ , c2∗ )
creasing in γ 2 . Further, g2 (γ2 ) = 0 when γ2 = λ − γ + . The revenue where c1∗ = c2∗ := c∗ . A natural scenario where such an equilibrium
maximization problem for the duopoly is re-stated as follows: is possible is when the two servers have identical delay functions.
In this section, we restrict to this case and assume that D j (· ) =
max R j (γ j ) = g j (γ j ) + c j − γj D(· ) for j = 1, 2. As the service systems are identical in their delay
γj
subject to 0 ≤ γ j ≤ γ j (c j − ) ≤ λ (P6 ) characteristics, it is desirable to identify conditions for existence of
given c j− . a symmetric Nash equilibrium. We begin with the following defi-
nition. Define α 1 , α 2 as follows.
For a given c j− , recall that γ j∗ (c j− ) denotes the maximizer of
program P4 and hence of the above program. Also recall that α1 = −γ + dg1γ(1γ1 )
Bˆ j (c j− ) denotes the best response admission price at Server j in
γ1 = γ + (16)
+ dg2 (γ2 )
response to the admission price c j− at the other facility. Then the α2 = −γ γ2
Nash equilibrium set of admission prices, denoted by (c1∗ , c2∗ ), is γ2 = γ +
characterized as follows: Based on these definitions, we have the following lemma. (See
Appendix for proof.)
(c1∗ , c2∗ ) = (c1 , c2 ) : Bˆ1 (c2 ) = c1 , Bˆ2 (c1 ) = c2 , (15)
Lemma 7. α1 = α2 .
where Bˆ j (c j− ) = g j (γ j∗ (c j− )) + c j− for j = 1, 2.
We begin the analysis for the duopoly problem by first identi- We now have the following theorem, that characterizes the nec-
fying that γ j∗ (c j− ) lies in the interior of the domain. We have the essary condition for a symmetric Nash equilibrium.
following lemma.
Theorem 2. Let c1∗ = c2∗ be a symmetric Nash equilibrium for the
γ j∗ (c j− ) / 0, γ j (c duopoly price competition. Then c1∗ = c2∗ = α1 .
Lemma 6. ∈ j− ) .
Proof. Recall that the Nash equilibrium is characterized as
Proof. See Appendix for proof.
(c1∗ , c2∗ ) = (c1 , c2 ) : Bˆ1 (c2 ) = c1 , Bˆ2 (c1 ) = c2 .
For a given c j− since γ j∗ (c j− ) lies in the interior of the domain,
dR j d2 R j
where Bˆ j (c j− ) = g j (γ j∗ (c j− )) + c j− for j = 1, 2. This implies that
γ j∗ (c j− )
satisfies
dγ j
= 0 and
dγ j2
≤ 0.
γ =γ ∗ γ j =γ j∗ c∗j = g j (γ j∗ (c∗j− )) + c∗j− and since c1∗ = c2∗ , we have g j (γ j∗ (c∗j− )) = 0
j j
dR d2 R for j = 1, 2. Now from Lemma 3 and symmetry of the servers, this
Define S j (c j− ) := γ j : dγ j = 0, 2j ≤ 0 . Then γ j∗ (c j− ) is ob- implies that γ1∗ (c2∗ ) = γ + and γ2∗ (c1∗ ) = λ − γ + . Since the servers
dγ j j
tained as a solution to the following. are identical, we have γ + = λ2 and hence γ2∗ (c1∗ ) = γ + . Since
γ j∗ (c j− ) is also a solution to program P8, γ j∗ (c j− ) ∈ S(c j− ). From
γ j∗ (c j− ) = arg max R j (γ j ). (P8 )
dR j
γ j ∈S j ( c j − ) the definition of S(c j− ), this implies that dγ j
= 0. Further,
γ j =γ +
From the above discussion, it should be clear that obtaining this implies from the definition of Rj (γ j ) that
the closed form expression for (c1∗ , c2∗ ) satisfying the simultaneous
equations of (15) is, in general, not easy. Note that our analysis dR j
dg j (γ j )
= γ+
+ g j (γ + ) + c∗j− (17)
until now makes minimal assumptions on the distribution func- d γ j γ j =γ + γj γ j =γ +
tion F or on the delay function Dj (·). For certain choices of these
= 0.
functions, it may be difficult to obtain a closed form expression for
γ j∗ (c j− ). The objective function Rj also need not be a concave func- We have g j (γ + ) = 0 and hence from the definition of α j for j =
tion. In that case, a brute force search among all the local maxima 1, 2 we have c∗j− = α j . From Lemma 7, we have α1 = α2 and hence
points needs to be carried out to choose the right γ j∗ (c j− ). c1∗ = c2∗ = α1 . This completes the proof.
Now recall from the introduction that the duopoly model with Note that the above theorem only provides a necessary condi-
heterogeneous customers has been studied numerically in Levhari tion for the Nash equilibrium pair and we shall soon see that in
and Luski (1978) when the delay function is of the M/M/1 type and fact this condition is not sufficient. We shall now provide a few
the delay cost distributions are Uniform and Pareto. More specifi- examples illustrating the occurrence of symmetric Nash equilibria.
cally, Levhari and Luski (1978) numerically demonstrate the exis- γ
tence and non-existence of Nash equilibrium prices under differ- Example 3. In this example, we assume that D j (γ j ) = μj for j =
j
ent parameter settings. The specific model considered in Levhari 1, 2. Let μ1 = μ2 = 4 while the arrival rate is λ = 3. We suppose
and Luski (1978) is a special case of our model setting since we al- that the distribution F(·) has a uniform distribution with support
low for more general delay functions and delay cost distributions. of [a, b]. We plot R1 (γ 1 ) and c1 (γ 1 ) as a function of γ 1 in Fig. 8.
In this paper, we do not indulge in performing such a numerical The aim of this example is to check whether (c1∗ , c2∗ ) = (α1 , α1 ) is
exercise since we do expect similar reaction curves (best response a symmetric Nash equilibrium. For the set of parameters of this
plots) for other distributions and delay cost functions. On the con- example we have γ + = 1.5 and since
trary, in the rest of this section, we shall explicitly characterize
dg1 (γ1 )
the symmetric Nash equilibrium prices for the system, something α1 = −γ +
which has not obtained until now. In the following subsection we γ1 γ1 = γ +
694 T. Bodas and D. Manjunath / European Journal of Operational Research 278 (2019) 686–698
Table 1 Table 2
The table indicates the price The estimates z can be
vector (c1 , c2 ), the measured obtained from Eq. (20)
value of γ 1 and the thresh- from the successive
old β obtained from Eq. (3). changes in the admis-
sion prices and the
c1 c2 γ1 β1 corresponding measure-
5.0 5 1.98 2.84 ments of the arrival
5.2 5 1.69 3.04 rates.
5.4 5 1.44 3.20
c1 c1 + δ z
5.6 5 1.23 3.33
5.8 5 1.05 3.44 5 5.2 0.37
6.0 5 0.89 3.53 5.2 5.4 0.39
6.2 5 0.75 3.60 5.4 5.6 0.41
6.4 5 0.63 3.67 5.6 5.8 0.42
6.6 5 0.52 3.37 5.8 6.0 0.44
6.8 5 0.43 3.78 6.0 6.2 0.44
6.2 6.4 0.45
6.4 6.6 0.46
6.6 6.8 0.47
The only unknown quantity is the exponential parameter α which
can now be obtained from the above equation.
The value of z for a fixed c1 and c1 + δ can be viewed as an esti- mission price and the arrival rate at Server 1 when c1 is decreased
mate for the density function f(x) and obviously z → f(x) as δ → 0. j times by δ , i.e., when c1 = c1 − jδ.γ1 =
implies that the most
j
These values of z for different values of c1 are given in Table 2. sensitive delay class β 1 must now be using Server 1 along with
696 T. Bodas and D. Manjunath / European Journal of Operational Research 278 (2019) 686–698
λ−γ1
F −1 λ = β1 (γ1 ) is decreasing in γ 1 . Clearly, g1 (γ 1 ) is mono- Proof. To reduce the notations, we represent γ j∗ (c j− ) by γ j∗ in
tone decreasing when γ 1 is such that 0 ≤ γ1 < γ +. the proof of the lemma. We shall prove that γ1∗ ∈
/ 0, γ 1 (c2 ) and
When γ 1 is such that γ + < γ1 ≤ λ, from the definition of γ + , the proof for γ2∗ ∈ / 0, γ 2 (c1 ) is along similar lines. Suppose γ1∗ ∈
we have (D2 (λ − γ1 ) − D1 (γ1 )) < 0. In this range of γ 1 , it can be {0, λ}. Then from the requirement that γ1∗ = λ − γ2∗ , we have ei-
seen from Eq. (8) that β 1 (γ 1 ) is continuous and increasing in γ 1 . ther (1) γ1∗ = 0 and γ2∗ = λ or (2) γ1∗ = λ and γ2∗ = 0. First consider
This again implies that g1 (γ 1 ) is continuous decreasing when γ 1 the case when γ1∗ = 0 and γ2∗ = λ. This implies that R1 (c1 (0 ), 0 ) =
satisfies γ + < γ1 ≤ λ. 0 and hence the revenue made by Server 1 at equilibrium is zero.
g1 (γ + ) = 0 follows from the definition of γ + where D1 (γ + ) = Further since this is an equilibrium, there is no incentive for the
D2 (λ − γ + ). The continuity at γ + is obvious from the fact that server to change the admission price and increase its revenue. We
g1 (γ + ) = 0 and limγ1 →γ + g1 (γ1 ) = 0. shall now show that this is not true. From Eq. (5) we know that
for a given c2 , the admission price at Server 1 must be at least
Lemma 4
c2 + g1 (0 ) > 0. Now we know that setting c1 = c2 will result in
Proof. Suppose ≥ g1 (0). From the definition of and from γ1 = γ + . Now due to the assumption that (1) D1 (0) < D2 (λ) and
Eq. (5), this implies that (2) D2 (0) < D1 (λ), there exists an
> 0 such that setting c1 = c2 +
698 T. Bodas and D. Manjunath / European Journal of Operational Research 278 (2019) 686–698
will result in γ1 ∈ (0, γ + ). The revenue earned is non-zero and Bodas, T., Ganesh, A., & Manjunath, D. (2014). Tolls and welfare optimization for mul-
there is clearly an incentive to deviate from any value greater than ticlass traffic in multiqueue systems arXiv:1409.7195.
Bodas, T., & Manjunath, D. (2011). On load balancing equilibria in multiqueue sys-
c2 + g1 (0 ). This implies that γ1∗ = 0 and γ2∗ = λ is not possible. The tems with multiclass traffic. In Proceedings of the NETGCOOP.
proof for γ1∗ = λ and γ2∗ = 0 is along the same lines. Bradford, R. M. (1996). Incentive compatible pricing and routing policies for multi-
-server queues. European Journal of Operational Research, 89, 226–236.
Lemma 7 Chen, H., & Frank, M. (2001). State dependent pricing with a queue. IIE Transactions,
33(10), 847–860.
Proof. From the definition of g1 (·) in Eq. (5) we have Chen, H., & Frank, M. (2004). Monopoly pricing when customers queue. IIE Transac-
tions, 36(6), 569–581.
dg1 (γ1 ) Chen, H., & Wan, Y. (2003). Price competition of make-to-order firms. IIE Transac-
= β1 (γ1 )(D2 (λ − γ1 ) − D1 (γ1 ) )
γ1 tions, 35(9), 817–832.
Dube, P., & Jain, R. (2008). N-player Bertrand-Cournot games in queues: Existence
+ β1 (γ1 ) D2 (λ − γ1 ) − D1 (γ1 ) of equilibrium. In Proceedings of the 46th annual allerton conference on commu-
nication, control, and computing, 2008 (pp. 491–498).
where the partial derivatives on the r.h.s. are w.r.t. γ 1 . Now from Dube, P., & Jain, R. (2010). Diffserv pricing games in multi-class queueing network
the definition of γ + and from Eq. (8) we have models. In Proceedings of the international teletraffic congress (ITC-22).