International Journal of Computer Science

and Engineering (IJCSE)

ISSN (P): 2278-9960; ISSN (E): 2278-9979
Vol. 7, Issue 4, Jun - Jul 2018; 1-10
© IASET

INVENTORY MODELS SOLVED WITH HYBRID APPROACHES

Xu-Ren Luo
Department of Computer Science and Information Engineering
Chung Cheng Institute of Technology, National Defense University, Taiwan, R.O.C.

ABSTRACT

This paper is a further discussion of models with ramp type or trapezoidal type demand. In a previously
published paper in European Journal of Operational Research that their solution procedure is accurate, but tedious that
will be substituted by our more compact approach to imply a board setting to solve a more general problem. We expand
their inventory model from a ramp type demand to an arbitrary demand to reveal a new discovery that the optimal solution
is not related to the demand type. Our paper will provide a new trend in inventory models to study new models and solution
procedures.

KEYWORDS:Ordering Cost, Inventory Model, Deteriorating Item

Article History
Received: 12 Oct 2017 | Revised: 09 Feb 2018 | Accepted: 26 May 2018

INTRODUCTION

Since Hill (1995) developed the inventory system with a ramp type demand there has been a trend towards
examining this kind of an inventory model both in depth and breadth. Some examples include Mandal and Pal (1998) with
deterioration items; Wu et al. (1999) assuming that the backlogging rate is proportional to the waiting time; Wu and
Ouyang (2000) with two different strategies beginning with stock or shortage; Wu (2001) with the Weibull distributed
deterioration; Giri et al. (2003) with a more generalized Weibull distributed deterioration; Deng (2005) to revise the work
of Wu et al. (1999); Manna and Chaudhuri (2006) to extend the inventory model a with time dependent deterioration rate;
Deng et al. (2007) to modify Mandal, and Pal (1998), Wu and Ouyang (2000).

However, their work is always restricted to ramp type demands. This paper points out that there is a general
property that is shared for every demand without being limited to the ramp type or any type of demand. Our findings show
that the lengthy discussion and the derivation of Deng et al. (2007) in the consideration of two different expressions of
ramp type demand are in fact unnecessary. There are several related papers that deserved to pay attention to them, for
examples, Chu et al. (1998), Chung et al. (1998), Lan et al. (1999), Chu and Chen (2001), Chu et al. (2004), Ronald et al.
(2004), Yang et al. (2005), Deng et al. (2006), Lin and Chu (2006), Chou et al. (2008), Deng et al. (2008), Liang et al.
(2008), Wu et al. (2008), Yang et al. (2011), Lin et al. (2012), Lin et al. (2013b). This paper attempts to investigate an
inventory model with stock in the first phase and then shortages in the second phase. In a multiple item inventory model, if
holding cost, deterioration cost and shortage cost are the same for all items, owing to the optimal solution is not related to
demand then different items with different demand still imply the same replenishment policy. Of particular interest is that

www.iaset.us editor@iaset.us
2 Xu-Ren Luo

when all targeted stock items reach zero simultaneously, and all subsequent demands are intentionally backlogged until the
end of the inventory cycle, then one-shot replenishment takes place. In so setting, the total inventory cost, including
holding cost, deteriorating cost, and shortage cost will achieve its minimum. As a matter of fact, this kind of design of an
inventory system has been prevalent in various industries such as the electronic industry, food industry, as well as the
financial industry. For instance, three major passive components: resistance, capacitance, and inductance in the IC industry,
the memory chip sets or IO (Input-Output) control device in IC sub-industries all refer to common materials. For years,
most common materials have been standardized in the specifications so as to be widely used in manufacturing electronic
components,semi-products or finished goods. In other words, under planned production or flexible manufacturing systems,
the average consumption rate for each of them can be accurately tracked, mastered, and modulated respectively. In this way,
the stock levels for these well-scheduled common materials will reduce to zero at the same time and hereupon the
reordering process and replenishment activities are automatically initiated.

ASSUMPTIONS AND NOTATION

We try to generalize the inventory model of Mandal and Pal (1998), Wu and Ouyang (2000) and Deng et al. (2007)
with the following assumptions and notations for the deterministic inventory replenishment policy with a general demand.

• The replenishment rate is infinite; thus, replenishments are instantaneous.

• The lead time is zero.

• T is the finite time horizon under consideration. Thus, it is a constant.

• C h is the inventory holding cost per unit per unit of time.

• C s is the shortage cost per unit per unit of time.

• C d is the cost of each deteriorated item.

• θ is the constant fraction of the on-hand inventory deterioration per unit of time.

• I (t ) is the on-hand inventory level at time t over the ordering cycle [0, T ] .

• The Shortage is allowed and fully backordered.

• The demand rate R (t ) is assumed to be any positive function for t > 0.

• t1 is the time when the inventory level reaches zero.

•
 C 
( )
f ( t1 ) is an auxiliary function defined as  C d + h  eθ t 1 − 1 − C s (T − t1 ) .
θ 


• C ( t1 ) is the total cost that consists of holding costs, deterioration cost, and shortage cost.

REVIEW OF PREVIOUS RESULTS

In Deng et al. (2007), they considered the ramp type demand where demand R (t ) is assumed as follows:

Impact Factor (JCC): 6.0127 NAAS Rating: 3.17

Inventory Models Solved with Hybrid Approaches 3

 D t, t <µ
R (t ) =  0 , (1)
D
 0 µ , t ≥ µ

so that µ is the changing point of linear demand to constant demand. Depending on the relation between µ and

t1 , they developed two inventory models: when µ ≥ t1 , the total cost, TC0 ( t1 ) , is defined for 0 ≤ t1 ≤ µ and for

µ < t1 , TC1 ( t1 ) is defined for µ ≤ t1 ≤ T . They studied properties of TC1 ( t1 ) in Theorem 1, and the properties of

TC0 ( t1 ) in Theorem 2. After which they found from Theorem 3, for the ramp type demand, that if ∆1 < 0 , the minimum

( )
#
occurs at TC1 t1 . On the other hand, if ∆1 ≥ 0 , then the minimum occurs at TC0 ( t1# ) where t1# is the unique solution

of f ( t1 ) = 0 for 0 ≤ t1 ≤ T with

 C 
( )
f (t1 ) =  Cd + h  eθ t 1 − 1 − C s (T − t1 ) ,
θ 
(2)


and ∆1 = f ( µ ) .

OUR PROPOSED INVENTORY MODEL

We consider an inventory model that starts with a stock. This model was first proposed by Hill (1995), and then
further investigated by Mandal and Pal (1998), Wu and Ouyang (2000) and Deng et al. (2007). Replenishment occurs at a
time t = 0 when the inventory level attains its maximum, S . From t = 0 to t1 , the inventory level reduces due to both
demand, R (t ) , and deterioration. At, t1 , the inventory level achieves zero, after which shortages are allowedduring the

time interval (t1 , T ) , and all of the demand during the shortage period (t1 , T ) is completely backlogged. The inventory

levels I (t ) of the model are described by the following equations:

d
I (t ) + θ I (t ) = − R (t ), 0 < t < t1 , (3)
dt
and

d
I (t ) = − R (t ), t1 < t < T . (4)
dt
We directly solve Equations (3) and (4) to imply that

t1

I ( t ) = e −θ t ∫ R ( x ) eθ x dx , for 0 ≤ t ≤ t1 , (5)
t

and

t1

I ( t ) = ∫ R ( x ) dx , for t1 ≤ t ≤ T . (6)
t

www.iaset.us editor@iaset.us
4 Xu-Ren Luo

The amount of deteriorated items during [0,t1 ] is evaluated

t1 t1

I ( 0 ) − ∫ R ( x )dx = ∫ R ( x ) ( eθ x − 1)dx . (7)

0 0

Using integration by part, the holding cost during [0,t1 ] is evaluated

eθ x − 1
t1 t1

Ch ∫ I ( t ) dt = Ch ∫ R ( x ) dx . (8)
0 0
θ

The shortage cost during [t1 , T ] is evaluated through integration by part

T T
Cs ∫ − I ( t )dt = Cs ∫ (T − x ) R ( x )dx . (9)
t1 t1

Therefore, the total cost is expressed as

eθ x − 1
t1 t1 T
C ( t1 ) = Cd ∫ R ( x ) ( eθ x − 1)dx +Ch ∫ R ( x ) dx +C s ∫ ( T − x ) R ( x )dx . (10)
0 0
θ t1

From Equation (10), it follows that

 C  
C ′ ( t1 ) = R ( t1 )  Cd + h  ( eθ t1 − 1) − Cs (T − t1 )  . (11)
 θ  

Motivated by Equation (11), we assume an auxiliary function, say f ( t1 ) ,

 C 
( )
f (t1 ) =  Cd + h  eθ t 1 − 1 − C s (T − t1 ) .
θ 
(12)


As a matter of fact, it is the same one proposed by Deng et al. (2007). By taking derivative of f ( t1 ) , i.e.,

f ′ ( t1 ) = (θ Cd + Ch ) eθ t1 + Cs > 0 , it is easy to find that f ( t1 ) increases from f ( 0 ) = −CsT < 0 to

 C 
f (T ) =  Cd + h  ( eθ T − 1) > 0 . Hence, obviously there exists a unique point, say t1# , that satisfies f ( t1# ) = 0 and
 θ 
the following equation holds


(
Ch  θ t 1#
 Cd +  e − 1 = Cs (T − t1 ) .
 θ 
#
) (13)

Since C ′ ( t1 ) ≤ 0 four 0 ≤ t1 ≤ t1# and C ′ ( t1 ) ≥ 0 for t1# ≤ t1 ≤ T such that t1# is the minimum solution for

C ( t1 ) . We summarize our findings in the next theorem.

Impact Factor (JCC): 6.0127 NAAS Rating: 3.17

Inventory Models Solved with Hybrid Approaches 5

Theorem 1 For the inventory model beginning with stock, the minimum solution satisfies f t1 = 0 and is not ( )
#

related to the demand.

DIFFERENT VIEWS OF OUR FINDINGS

We intend to provide a marginal (cost) analysis to discuss our findings. If there is an item in demand quantity
R (t )dt where dt is a small (infinitesimal) time interval for the demand during [t , t + dt ] , then there are two
replenishment policies: (a) fulfill the demand for the stock, or (b) satisfy the demand from backorder. If we decide to fulfill

the demand for the stock, we need to store R(t )e dt at time

θt d
t = 0 , since the solution of I (t ) + θ I (t ) = 0 is
dt
I ( t ) = I ( 0 ) e−θ t so that the beginning stock is I (0 ) = R(t )eθ t dt . It follows that after deteriorated items are removed,

the remaining stock R (t )dt is just enough to meet the demand. The amount of deteriorated items is R (t ) e ( θt
)
− 1 dt .

From the inventory level R (t )e e dt for, x ∈ [ 0, t ] , the holding cost can be calculated by
θt −θ x

eθ t − 1
t
C h ∫ R (t )dte θ t e −θ x dx = C h R (t )dt . (14)
0
θ

Hence, the total cost for demand R (t )dt fulfilled from the stock is
 Ch


( )
+ C d  R (t )dt e θ t − 1 . On the other
θ 
hand, if the policy of backlog is adopted then the shortage cost is R (t )dtC s (T − t ) . We find that if

 Ch


( )
+ C d  R (t )dt e θ t − 1 < R(t )dtC s (T − t ) , (15)
θ 

then the better policy is to satisfy demand from the stock. Otherwise, if

 Ch


( )
+ C d  R (t )dt e θ t − 1 > R(t )dtC s (T − t ) . (16)
θ 

Then the better policy is to hold demand as backlog until the replenishment takes place. Recall from Equation (12),

 C 
we have shown that
 θ 
( )
f ( t ) =  Cd + h  eθ t − 1 − Cs (T − t ) has a unique root, say t1# . It means that for those

demands occurring during  0, t1

#
) , they should be fulfilled from the stock And for those demands occurring during
(t#
1 , T  , they should be replenished by backorder.Our marginal (cost) analysis comes up with the same results as our
previous analytical approach in Section 4.

NUMERICAL EXAMPLES

First, we recall the numerical examples in Deng et al. (2005) with C d = 5 , C h = 3 , C s = 15 , D0 = 100 ,

www.iaset.us editor@iaset.us
6 Xu-Ren Luo

θ = 0.001 , β = 2 and T = 1 . For their example 1, µ = 0.12 , and for their example 2, µ = 0 .9 . After a lengthy

computation, Deng et al. (2005) derived that t1 = 0.8331 for their examples 1. By the similar complicated evaluation,
*

Deng et al. (2005) obtained that t1 = 0.8331 for their examples 2. We can claim that Deng et al. (2005) did not aware
*

that the optimal solution is not related to the demand type such that their example 2 should have the optimal solution as
their example 1.

For the second example, we consider the numerical examples in We et al. (2008) with the following data:

C d = 0 , C h = 3 , C s = 5 , D0 = 400 , θ = 0.05 , T = 1 , and µ = 0.4 , 0.5953 and 0.6 for three different cases.

Wu et al. derived that the optimal solution t1 = 0.5953 ,

*
0.5953 and 0.5953 for three different cases. Their optimal
solutions are consistent with our theorem that the optimal solution is not related to the demand type such that three
different values of µ will result in the same optimal solution.

For the third numerical examples, we refer to Deng et al. (2007) with the following data: C d = 5 , C h = 3 ,

C s = 15 , D0 = 100 , θ = 0.001 , T = 1 , and µ = 0.12 , 0.9 and 0.8 for three different cases. Deng et al. (2007)

derived that the optimal solution t1 = 0.833044 ,

*
0.833044 and 0.833044 for three different cases. From the above
three examples, it indicates that Deng et al. (2007) did not realize that the optimal solution is not related to the demand
such that to provide three numerical examples is unnecessary.

For the fourth example, we consider Yang et al. (2011) with C1 = 1 , C 2 = 25 , C 3 = 3 , C 4 = 25 , α = 4,

δ =4 and T = 1 . They constructed three examples with different demand type: (a) R (x ) = 100 x , for 0 ≤ x ≤ 0.12
and R (x ) = 12 , for 0.12 ≤ x ≤ 1 , (b) R ( x ) = 20 − 5 x , for 0 ≤ x ≤ 1 , and (3) R ( x ) = 20 + 5 x , for 0 ≤ x ≤ 1 . For
their three different demands, Yang et al. (2011) derived the same optimal solution t1 = 0.678 . It indicates Yang et al.
*

(2011) did not know that the optimal solution is not related to the demand type.

For the fifth example, we recall Skouri et al. (2009). They assumed that µ = 0.12 for their example 1 and

µ = 0 .9 for their example 2. They derived that t1 = 0.8604 >

*
µ for their example 1 and t1 = 0.8604 <
*
µ for their

example 2. From their dissertation of “ t1 = 0.8604 >

*
µ for their example 1 and t1 = 0.8604 <
*
µ for their example 2”,
we can conclude that they did not know that the optimal solution is not related to the demand type.

For our sixth example, we consider Tung (2013). He pointed out that Deng et al. (2007) obtained

t1* = 0.833044 for three different examples with µ = 0.12 for their example 1, µ = 0.9 for their example 2, and
µ = 0 .8 for their example 3. Tung (2013) mentioned that to provide a reasonable explanation for this coincide

phenomenon will be an interesting research topic. This paper provide an positive solution for his question: the optimal
solution is not related to the demand such that three different examples with different demand will have the same optimal
solution.

Impact Factor (JCC): 6.0127 NAAS Rating: 3.17

Inventory Models Solved with Hybrid Approaches 7

DIRECTION FOR FUTURE RESEARCH AND APPLICATION FOR OUR FINDINGS

There are several papers: 0Chung et al. (2000), Chu and Chen (2002), Chu et al. (2003), Chu et al. (2005), Deng
et al. (2005), Deng et al. (2007), Yang et al. (2008), Chou et al. (2009), Lin et al. (2011), Lin et al. (2013a), and Lin et al.
(2014) that only used analytic methods to locate the optimal solution. Following our managerial point of view without
constructing a system of differential equations may be a possible alternative solution approach for papers. It will help
ordinary readers realize inventory models from a different aspect.

CONCLUSIONS

In the paper, we discover an interesting phenomenon for the finite time horizon inventory model to show that the
optimal solution is not related to the demand. The generalized form of inventory model is developed and the corresponding
optimal solution is derived. We believe that our findings provide an essential benchmark for those researchers who have
the motive of pursuing different optimal inventory systems along with the changes of demand types. This study gives the
solid evidence that the optimal solution is not related to the demand no matter ramp type, trapezoid type, fixed type or any
other kinds of it are targeted.

REFERENCES

1. Chou, S., Chouhuang, W. T., Lin, J. S., & Chu, P. (2008). An analytic solution approach for the economic order
quantity model with Weibull ameliorating items. Mathematical and Computer Modelling, 48(11-12), 1868-1874.

2. Chou, S., Julian, P. C., & Hung, K. (2009). A note on fuzzy inventory model with storage space and budget
constraints. Applied Mathematical Modelling, 33(11), 4069-4077.

3. Chu, P., & Chen, P. S. (2001). A note for "on an inventory model for deteriorating items and time-varying
demand". Mathematical Methods of Operations Research, 53(2), 297-307.

4. Chu, P., & Chen, P. S. (2002). A note on inventory replenishment policies for deteriorating items in an
exponentially declining market. Computers and Operations Research, 29(13), 1827-1842.

5. Chu, P., Chen, P. S., & Niu, T. (2003). Note on supplier-restricted order quantity under temporary price
discounts. Mathematical Methods of Operations Research, 58(1), 141-147.

6. Chu, P., Chung, K., & Lan, S. (1998). Economic order quantity of deteriorating items under permissible delay in
payments. Computers and Operations Research, 25(10), 817-824.

7. Chu, P., Yang, K., & Chen, P. S. (2005). Improved inventory models with service level and lead time. Computers
and Operations Research, 32(2), 285-296.

8. Chu, P., Yang, K., Liang, S., & Niu, T. (2004). Note on inventory model with a mixture of back orders and lost
sales. European Journal of Operational Research, 159(2 SPEC. ISS.), 470-475.

9. Chung, K., Chu, P., & Lan, S. (2000). Note on EOQ models for deteriorating items under stock dependent selling
rate. European Journal of Operational Research, 124(3), 550-559.

10. Chung, K., Lin, S., Chu, P., & Lan, S. (1998). The production and inventory systems taking account of time
value.Production Planning and Control, 9(6), 580-584.

www.iaset.us editor@iaset.us
8 Xu-Ren Luo

11. Deng, P. S., (2005) Improved inventory models with ramp type demand and Weibull deterioration. International
Journal of Information and Management Sciences, 16 (4), 79-86.

12. U.K.Bera, J.K.Dey & M.Maiti, Inventory Model With Fuzzy Lead-Time and Dynamic Demand Over Finite Time
Horizon Using Interactive Fuzzy Method, International Journal of Applied Mathematics & Statistical Sciences
(IJAMSS), Volume 1, Issue 1, August-September 2012, pp. 50-72

13. Deng, P. S., Yang, G. K., Chen, H., Chu, P., & Huang, D. (2005). The criterion for the optimal solution of
inventory model with stock-dependent consumption rate. International Journal of Information and Management
Sciences, 16(2), 97-109.

14. Deng, P. S., Yen, C., Tung, C., Yu, Y., & Chu, P. (2006). A technical note for the deteriorating inventory model
with exponential time-varying demand and partial backlogging. International Journal of Information and
Management Sciences, 17(2), 101-108.

15. Deng, P. S., Lin, R., and Chu, P. (2007) A note on the inventory models for deteriorating items with ramp type
demand rate. European Journal of Operational Research, 178, 112-120.

16. Deng, P. S., Lin, R. H., Hung, C., Chang, J., Chao, H., Jung, S., Chu, P. (2008). Inventory models with a negative
exponential crashing cost taking time value into account. Journal of the Operations Research Society of Japan,
51(3), 213-224.

17. Giri, B. C., Jalan, A. K., and Chaudhuri, K. S. (2003) Economic Order Quantity model with Weibull deterioration
distribution, shortage and ramp-type demand. International Journal of Systems Science 34 (4), 237-243.

18. Hill, R.M., (1995) Inventory models for increasing demand followed by level demand. Journal of the Operational
Research Society, 46 (10), 1250-1259.

19. Lan, S. P., Chu, P., Chung, K., Wan, W. J., & Lo, R. (1999). A simple method to locate the optimal solution of the
inventory model with variable lead time. Computers and Operations Research, 26(6), 599-605.

20. Liang, S., Chu, P., & Yang, K. (2008). Improved periodic review inventory model involving lead-time with
crashing components and service level. International Journal of Systems Science, 39(4), 421-426.

21. Lin, J., Chao, H., & Julian, P. (2013a). A demand independent inventory model. Yugoslav Journal of Operations
Research, 23(1), 129-135.

22. Lin, J., Chao, H. C. J., & Julian, P. (2013b). Planning horizon for production inventory models with production
rate dependent on demand and inventory level. Journal of Applied Mathematics, 2013

23. Lin, J., Chao, H. C., & Julian, P. (2012). Improved solution process for inventory model with ramp-type demand
under stock-dependent consumption rate. Journal of the Chinese Institute of Industrial Engineers, 29(4), 219-225.

24. Lin, J., Hung, K., & Julian, P. (2014). Technical note on inventory model with trapezoidal type demand. Applied
Mathematical Modelling, 38(19-20), 4941-4948.

25. Lin, R. H., & Chu, P. (2006). Note on stochastic inventory models with service level constraint. Journal of the
Operations Research Society of Japan, 49(2), 117-129.

Impact Factor (JCC): 6.0127 NAAS Rating: 3.17

Inventory Models Solved with Hybrid Approaches 9

26. Lin, S., Wou, Y., & Julian, P. (2011). Note on minimax distribution free procedure for integrated inventory model
with defective goods and stochastic lead time demand. Applied Mathematical Modelling, 35(5), 2087-2093.

27. Mandal, B. and Pal, A.K. (1998) Order level inventory system with ramp type demand rate for deteriorating items.
Journal of Interdisciplinary Mathematics, 1, 49-66.

28. Manna, S. K., Chaudhuri, K. S., (2006) An EOQ model with ramp type demand rate, time dependent deterioration
rate, unit production cost and shortages. European Journal of Operational Research 171 (2), 557-566.

29. Ronald, R., Yang, G. K., & Chu, P. (2004). Technical note: The EOQ and EPQ models with shortages derived
without derivatives. International Journal of Production Economics, 92(2), 197-200.

30. Skouri, K., Konstantaras, I., Papachristos, S., Ganas, I., 2009. Inventory models with ramp type demand rate,
partial backlogging and Weibull deterioration rate. European Journal of Operational Research 192 (1), 79–92.

31. Tung C. T. (2013): A note on the inventory models with ramp type demand of negative exponentially distributed
changing point, Journal of Industrial and Production Engineering, 30:1, 15-19

32. Wu, J. K., Lei, H., Jung, S., & Chu, P. (2008). Newton method for determining the optimal replenishment policy
for EPQ model with present value. Yugoslav Journal of Operations Research, 18(1), 53-61.

33. Wu, J.W., Lin, C., Tan, B., and Lee, W.C. (1999) An EOQ model with ramp type demand rate for items with
Weibull deterioration. International Journal of Information and Management Sciences, 10, 41-51.

34. Wu, K.S., and Ouyang, L.Y. (2000) A replenishment policy for deteriorating items with ramp type demand rate,
Proceeding of National Science Council ROC (A), 24, 279-286.

35. Wu, K.S. (2001) An EOQ inventory model for items with Weibull distribution deterioration, ramp type demand
rate and partial backlogging, Production Planning & Control, 12, 787-793.

36. Wu K.S, Ouyang L.Y., Yang C.T. (2008). Retailer’s optimal ordering policy for deteriorating items with ramp-type
demand under stock dependent consumption rate. International Journal of Information and. Management.
Science, 19, 245-262.

37. Yang, G., Ronald, R. J., & Chu, P. (2005). Inventory models with variable lead time and present value. European
Journal of Operational Research, 164(2), 358-366.

38. Yang, G. K., Chou, S., Hung, C., Lin, J. S., & Chu, P. (2008). Research note on the criteria for the optimal solution
of the inventory model with a mixture of partial backordering and lost sales. Applied Mathematical Modelling,
32(9), 1758-1768.

39. Yang, G. K., Lin, R., Lin, J., Hung, K., Chu, P., & Chouhuang, W. (2011). Note on inventory models with Weibull
distribution deterioration. Production Planning and Control, 22(4), 437-444.

www.iaset.us editor@iaset.us