CONTRACT MECHANISMS FOR REVENUE SHARING IN A SUPPLY CHAIN

Dong Won Cho

1
, Young Hae Lee
*2
, Se Ho Park
3
, Mitsuo Gen
4

1-4
Dept. of Industrial & Management Engineering
Hanyang University, South Korea
4
Fuzzy Logic Systems Institute, Iizuka, Japan
dwcmjcho@hanyang.ac.kr
1
, yhlee@hanyang.ac.kr
*2
, scm@hanyang.ac.kr
3
,mitsuogen@gmail.com
4


Abstract:
One of the most important things in pursuit of supply chain management is to prevent sub-optimization
caused by decentralized decision making over the various entities. As a solution of the issue, supply chain
coordination approach has frequently been used. However, a coordinated supply chain might fail to
provide additional profit to one of the players. A supply chain contract can be to achieve the same profit
of supply chain coordination and to improve the benefit of all entities involved. In this paper, we address
the model of supply chain contract, which is a combination contract of a quantity discount contract based
on a revenue sharing contract, which is used to coordinate under a multi-echelon supply chain facing a
stochastic customer demand. It is shown that the proposed contract can achieve supply chain coordination
and win-win situation.

Keywords: Supply chain management, Revenue sharing, Coordination, Contract, Win-Win


1. Introduction
A well-designed supply chain contracts are a useful tool to improve the performance of all entities
included as well as prevent sub-optimization. A self-serving focus of the supply chain entities often results
in sub-optimization with poor systemwide performance. A traditional solution to the issue is the use of
supply chain coordination. To coordinate a supply chain, a centralized or decentralized decision-making
approach can be adopted. The former option occurs when all decisions are made by a single entity in the
supply chain, the latter when several independent entities make decision at the different supply chain
stages. However, although a coordinated supply chain can achieve the best performance through
optimizing the entire supply chain, win-win, which occurs if every supply chain entities make higher
profit compared to the decentralized decision making situation, is not necessarily obtained. Furthermore,
even if all the entities gain from their collaboration, the supply chain is necessarily optimized. That is, it is

*
Corresponding Author

Proceedings of the 41st International Conference on Computers & Industrial Engineering
599

to be noted that a coordinated supply chain does not imply win-win. When all supply chain entities is
independent and autonomous organizations, clearly, from a practical point of view, win-win is probably
more important than supply chain coordination. After all, the entity is only prompted to participate in joint
actions if he will obtain a profit from the collaboration. As a useful means to achieve supply chain
coordination and win-win, supply chain contracts is frequently adopted.
Most of literature on supply chain contracts focused on a situation with only two entities: a buyer and
a supplier (Cachon, 2003). These models include revenue sharing (Cachon and Lariviere, 2000), quantity
discounts (Burnetas et al., 2007), buy back (Emmons and Gilbert, 1998). quantity flexibility (Tsay, 1999).
The focus of this paper is on a combination contract of a Quantity Discounts (QD) contract based on a
Revenue Sharing (RS) contract with demand uncertainty. Under a RS contract, the buyer shares some of
its revenue with the supplier. A RS contract thus involves two parameters, namely the wholesale price per
unit and a percentage of the buyer's revenue that goes to the supplier (Koulamas, 2006; Lee and Whang,
1999). Under a QD contract, a supplier who offers the buyer a quantity discount varies the price charged
to the buyer retailer according to the quantity purchased by the retailer. The buyer obtains a discount for
purchasing a larger quantity of the product from the supplier. A QD involve two parameters, with a fixed
base price and an additional price which is a decreasing function in an order quantity (Burnetas et al.,
2007). In result, a combination contract discussed in this paper can involve four parameters.
The literature on a multi-echelon supply chain contract has recently proposed two contract models
called as pairwise contract and spanning contract. The former is to install contracts between all pairs of
adjacent entities in a multi-echelon supply chain (Giannoccaro and Pontrandolfo, 2004), while the latter is
that one supply chain entity takes the lead in negotiating a single contract with all other entities
simultaneously (Van de rhee et al., 2010). In the previous research, the pairwise contract may be unlikely
to occur in the whole multi-echelon supply chain. The main reason of this result is because it implicitly
assumes that all contracts between the pairs of entities are installed simultaneously. However, spanning
contract can overcome this drawback of pairwise contract because a single entity initiates a single contract
with all other entities simultaneously. Van de rhee et al. (2010) have proposed the spanning RS contract
where the most downstream entity initiates a single contract involving all upstream entities with
considering a practical point of view.
This paper deals with a combination contract of a QD contract based on a RS contract, which is used to
coordinate under a multi-echelon supply chain facing a stochastic customer demand. We call this type of
RS contract the spanning RSQD contract. We focus on supply chain contracts addressing supply chain
coordination, namely the spanning RSQD contract. Furthermore, particular attention is devoted to support
the fine tuning of the contract parameters so as to achieve a winwin situation. The fundamental idea
underlying the spanning RSQD contract is that if each entity decreases its wholesale price, their respective
buyers have an incentive to decrease their prices also. As a consequence, the most downstream entity will
motivate him to order more. This in turn increases availability to the final customer and therefore
Proceedings of the 41st International Conference on Computers & Industrial Engineering
600

positively influences the overall supply chain revenue. Clearly, the various entities might want to receive
some compensation for their decreased prices, which in the spanning contract is given in terms of a share
of the revenue of the most downstream entity.
The paper is organized as follows. In Section 2, we introduce the basic supply chain model and discuss
the spanning RSQD contract. In Section 3, necessary and sufficient conditions for the win-win solutions
under the spanning RSQD contract are given. Finally, in Section 4 the study is concluded.

2. The multi-echelon supply chain model
The basic model setting introduced in this paper is similar to the one described in Van de rhee et al.
(2010). In this model, the supply chain is a simple serial structure with a single entity at each of the
n 2 echelons. Denote Entity 1 as the most downstream and Entity n as the most upstream organization.
Let Entity i buy a single product at purchasing cost W
i+1
and operational costs c
i
per unit and sell it to
Entity i - 1 at W
i
, i = 2, , n. For notational convenience, it is assumed that W
n+1
= 0. Thus, Entity 1
faces a purchasing cost W
2
and operational costs c
1
per unit and sells it to the end-customers at p per unit
and can decide on the order quantity Q. Since the market price is higher than the total supply chain cost, it
is assumed that p > c = c

n
=1
, where the cost c
i
can represent the operational costs and/or the shipping
costs at the various entities. Furthermore, based on the spanning contract, it is assumed that all entities
have full information on all data. For that reason, no further assumptions on the supply chain cost structure
are made.
In a market-like setting, the system would work as follows. All entities face the newsvendors problem.
Therefore, the product has a short life cycle and there is only a one-time order. Moreover, since the
product has a short life cycle, there is no stock from previous periods, and any unused stock from previous
periods cannot be used for future periods. Furthermore, it is assumed that each entity has the objective to
maximize its own expected profit. The order quantity Q is passed on through the entire supply chain. And
it is assumed that each entity has sufficient capacity to handle any realistic order size and any unmet
demand is lost as reordering is not possible. Let > u be end-customer demand during the selling
season. Let F

() be the distribution function of end-customer demand and

() its probability
density function: F

() is differentiable, strictly increasing and F

(u) = u.
The following sequence of events within this supply chain occurs. First Entity n determines its selling
price W
n
, followed by Entity n - 1, all the way to Entity 2 that determines its price W
2
. Given W
2
, c
1
and
the demand distribution

(), Entity 1 submits an order quantity Q to Entity 2. The same order quantity
is passed on from Entity 2 to Entity 3, all the way to Entity n. Subsequently, the amount Q is shipped from
Entity n all the way to Entity 1 before the selling season. Finally, the customer demand D occurs and
Entity 1 sells the amount Min{Q, D} to the end-customer. Let F

() = ]

()

0
J, and 0

() =
]

()

0
J. The expected profit of Entity 1 is n
1
= (p - w
2
- c
1
) - pF

() + p0

(). The
Proceedings of the 41st International Conference on Computers & Industrial Engineering
601

expected profit of Entity 2, , n is n

= (w

- w
+1
- c

). And the expected profit of the total the

supply chain is n
sc
= n

n
=1
= (p - c) - pF

() + p0

().
To achieve supply chain coordination, it is assumed that a centralized control involves the existence of a
unique decision-maker in the supply chain. Therefore, all entities in supply chain work together as a single
entity with the objective to maximize the supply chain profit. Clearly, the centralized supply chain (which
is denoted by superscript c), is facing the well-known newsvendor problem with per-unit overstocking
cost c and the per-unit understocking cost p - c . The supply chain optimal order quantity
c
is

c
= (F

)
-1
[
p-c
p
. The associated profit is n
sc
c
= p0

(
c
).
Now consider the decentralized supply chain (which is denoted by superscript d), in which it is assumed
that there is no coordination, i.e., all entities act independently and take decisions that maximize their
respective profits. Although there are various approaches possible to analyze this situation, it will be
assumed that optimization takes place in a Stackelberg sequence proposed by Van de rhee et al. (2010),
where in the time-line Entity n is the first decision maker, and the model is solved through backwards
induction. Thus, first he optimal order quantity
d
of Entity 1 is determined (as a function of w
2
d
), then
the optimal w
2
d
for Entity 2 is decided (as a function of w
3
d
), all the way to Entity n determining its
optimal w
n
d
. In the decentralized scenario, the expected profit of Entity 1 is n
1
d
= (p - w
2
d
- c
1
) -
pF

() + p0

(). Entity 1 faces a newsvendor problem with per-unit overstocking cost c

1
+ w
2
d
and
the per-unit understocking cost p -c
1
+w
2
d
. The supply chain optimal order quantity is
d
=
(F

)
-1
_
p-w
2
d
-c
1
p
]. The expected profit of Entity 1 is n
1
d
= p0

(
d
). The expected profit of Entity
2, , n is n

d
= (w

d
-w
+1
d
- c

)
d
. And the expected profit of the total the supply chain is
n
sc
d
= n

n
=1
= (w
2
d
+ c
1
- c)
d
+ p0

(
d
). Note that, unlike in the centralized supply chain, here
the supply chain profit depends on the wholesale price w
2
d
. According to Van de rhee et al. (2010)
observations, for similar situation, the decentralized supply chain results in a suboptimal supply chain
profit, i.e., n
sc
c
> n
sc
d
.
The spanning RSQD contract (denoted by superscript sq) is a combinational contract of a quantity
discount based on a revenue sharing. This contract is primarily governed by the wholesale prices w

sq

and the percentages

sq
of the revenues of Entity 1 that are shared with Entity i = 2, , n (with
u =

sq
n
=2
1). And all entities fix a base price y

sq
and charge an additional price p

sq
which
is a decreasing function in Q. At its simplest form the price at which the supplier sells the product to the
buyer is dependent upon the order quantity. The price of Entity 2, , n is w

sq
= y

sq
+

i
sq

. The
expected profit of Entity 1 is n
1
sq
= _(1 - )p - y
2
sq
-

2
sq

- c
1
_ -(1 -)pF

() +(1 -
Proceedings of the 41st International Conference on Computers & Industrial Engineering
602

)p0

(). The expected profit of Entity 2, , n is n

sq
= _

sq
p + y

sq
+

i
sq

-y
+1
sq
-

i+1
sq

- c

] -

sq
pF

() +

sq
p0

().
Theorem 1. The spanning RSQD contract (y
2
sq
, . . . , y
n
sq
; p
2
sq
, . . . , p
n
sq
;
2
sq
, . . . ,
n
sq
) with

y
2
sq
= (1 -)c - c
1
(1)

coordinates the supply chain.
Proof. For the spanning RSQD contract, Entity 1 would choose Q to optimize its profit as given by n
1
sq
.
As Entity 1 faces a newsvendor problem with per-unit overstocking cost y
2
sq
+ c
1
and the per-unit
understocking cost (1 -)p - y
2
sq
-c
1
, the optimal order quantity is
sq
= (F

)
-1
_
(1-q)p-y
2
sq
-c
1
(1-q)p
].
Under condition (1) it follows that
sq
=
c
. Using the optimal order size, the expected profit of Entity 1
is n
1
sq
= -p
2
sq
+ (1 -)p0

(
sq
) = -p
2
sq
+ (1 -)n
sc
c
. Furthermore, using condition (1), the profits
for Entity i = 2, , n are n

sq
= _

sq
c + y

sq
+

i
sq

sq
- y
+1
sq
-

i+1
sq

sq
- c

]
sq
+

sq
n
sc
c
. And, again
condition (1), the total supply chain profit is n
sc
sq
=

sq
n
=1
= n
sc
c
+ c
sq
+ ((1 - )c - c
1
)
sq
-
c

sq n
=2
= n
sc
c
. Therefore, the maximal supply chain profit is achieved.

3. Win-Win Solutions
Note that in order to achieve supply chain optimization under the RSQD contract, only a condition is
needed on how y
2
sq
is related to

sq
(i = 2, , n). Furthermore, note that there is still considerable
freedom to choose y
2
sq
,
2
sq
, . . . ,
n
sq
. These observations are used to derive win-win solutions.
Clearly, under the assumption that condition (1) is fulfilled, which means that the spanning RSQD
contract coordinates the supply chain, the necessary conditions for achieving a win-win outcome are

n
1
sq
= -p
2
sq
+ (1 -)n
sc
c
> n
1
d
, (2)

for i = 2, since y
2
sq
is determined by condition (1)

n
2
sq
= _(
2
s
+1 -)c - c
1
+
p
2
sq

sq
-y
3
sq
-
p
3
sq

sq
- c
2
_
c
+
2
sq
n
sc
c
> n
2
d
, (S)

and for i = 3, . . . , n,

Proceedings of the 41st International Conference on Computers & Industrial Engineering
603

n

sq
= _

sq
c + y

sq
+
p

sq

sq
- y
+1
sq
-
p
+1
sq

sq
-c

_
sq
+

sq
n
sc
c
> n

d
. (4)

Let 0 denote the set of all RSQD contract (y
2
sq
, . . . , y
n
sq
; p
2
sq
, . . . , p
n
sq
;
2
sq
, . . . ,
n
sq
) that
satisfy the above conditions (2)-(4). Clearly, all RSQD contracts in 0 achieve both coordination and a
win-win outcome.

4. Conclusion
This paper has addressed supply chain contract model, which is a combination contract of a quantity
discount contract based on a revenue sharing contract. In particular, a combination contract model has
been proposed to coordinate a multi-echelon supply chain. It has been shown that a combination contact
lets all entities select order quantities that are optimal for the whole supply chain. Therefore, the model
guarantees supply chain coordination. Furthermore, knowing that a winwin condition is really achievable
could be a good incentive to the collaboration required in the contract design.

References
Burnetas, A., Gilbert, S. M., and Smith, C. E. (2007), Quantity discounts in single-period supply contracts
with asymmetric demand information, IIE Transactions, Vol. 39, pp. 465-479.
Cachon, G. P. (2003), Supply chain coordination with contracts, de Kok A. G. and Graves, S. C., Eds.,
Supply chain management: Design, coordination and operations handbook, Elsevier, Amsterdam. pp.
229-339.
Cachon, G. P., and Lariviere, M. A. (2005), Supply chain coordination with revenue sharing contracts:
strengths and limitations, Management Science, Vol. 51, pp. 30-44.
Emmons, H., and Gilbert, S. M. (1998), The role of returns policies in pricing and inventory decisions for
catalogue goods, Management Science, Vol. 44, pp. 276-283.
Giannoccaro, I., and Pontrandolfo, P. (2004), Supply chain coordination by revenue sharing contracts,
International Journal of Production Economics, Vol. 89, pp. 131-139.
Koulamas, C. (2006), A newsvendor problem with revenue sharing and channel coordination, Decision
Sciences, Vol. 37, pp. 91-100.
Lee, H., and Whang, S. (1999), Decentralized multi-echelon supply chains: incentives and information,
Management Science, Vol. 45, pp. 633-640.
Tsay, A. (1999), The quantity flexibility contract and supplier-customer incentives, Management Science,
Vol. 45, pp. 1339-1358.
Van der Rhee, B., Van der Veen, J. A. A., Venugopal, V., and Nalla V. R. (2010), A new revenue sharing
mechanism for coordination multi-echelon supply chains, Operations Research Letters, Vol. 38, pp.
296-301.
Proceedings of the 41st International Conference on Computers & Industrial Engineering
604