Proceedings of the Asia Pacific Industrial Engineering & Management Systems Conference 2015

Supply Chain Design for Vietnam Agriculture Product:

A Case Study of Milk-apple
Bui Duy Tan†
Department of Industrial & Systems Engineering
International University – Vietnam National University, Ho Chi Minh city, Vietnam
Tel: (+84) 993952706, Email: [email protected]

Ho Thanh Phong
Department of Industrial & Systems Engineering
International University – Vietnam National University, Ho Chi Minh city, Vietnam
Tel: (+84) 903718904, Email: [email protected]

Abstract. With the tradition of agriculture, Vietnam has long been regarded the land of many special tropical
fruits. However, the supply chain for such fruits are now facing with various problems which lead to
considerable diminishing in the profit. In this paper, taking the milk-apple in Vietnam as a case study, the fruit
is now facing with the problems of big fluctuate demand, storing method and ineffectively working supply
chain network. Many techniques such as: demand analysis and forecasting, products design, supply chain
network design, facilities planning, and so on will be used to solve the current difficulties. After the study, it is
realized that the local authority policy should take into consideration for the profit of the proposed
improvements in the paper and conduct to implement it. However, due to lacking of practical and limitation of
time, the marketing research and some other aspects are not covered and need to be considered in further
study.

Keywords: Demand Forecasting, Supply Chain Network Design, Facilities Planning

1. INTRODUCTION facing with the rapidly decreasing of price due to the

current ineffectively operating supply chain network. All of
Vietnam has long been well-known as the land of the procurement hubs (the places that the farmers sell their
agriculture. According to Vietnam General Statistic Office, fruits to the dealers), the warehouses are just located
the total area of agricultural land account for 79.4% of the without any calculation by the local people. As a result, the
country’s total land area in 2013. There are many special fruit was transported from many places to many other
agriculture products in Vietnam such as: rice, dragon fruit, places and finally to the assembly point before transporting
mango, durian, madarin and others, which long time bring to big cities. Moreover, the current method of storing the
benefits to the country by adapting not only inland demand fruit and the way of the farmer controlling the productivity
but exotic demand also. to adapt customer demand is affecting negatively to the
This paper will take a case study of Vietnam milk- benefit of the fruit. The milk-apple has the harvesting
apple, which is a special fruit with good quality and season from December to May and there is no product in
exclusive flavor with all of the other kinds of milk-apple all the remaining period. The farmer in here cannot forecast
over the world. The fruit is grown mainly in the Mekong their customer demand and they just estimating and
delta - south of the country. According to the statistics from harvesting as soon as the fruit is ripen enough, supply to
the official website of the People’s Committee of Tien their customer without knowing if their customers need or
Giang province – Vietnam, in 2004, the area has a total of not. Consequently, they will over-supply the product to
about 2 thousand ha of growing the fruit, contributing for their customers in the season months and cannot adapt
more than 30 thousand tons of productivity per year and customer demand in the out-of-season months. Furthermore,
has long been bringing a big profit for the farmers. the Vietnam milk-apple has a very short lifecycle – 2 days
However, for the nearest ten years, the fruit has been from harvesting and up to now, there is no standard

†: Corresponding Author

1129
 Tan and Phong

warehouse system for storing the fruit in the area. In many and Peter Meindl (2007), in their work ‘Supply Chain
cases, if the harvesting day is coming and the customer Management, Strategy, Planning and Operation’, indicated
demand fluctuates, the farmers cannot adapt the situation. that, facility location decisions have a long-term impact on
As a result, the dealer has to buy the fruit from the farmers a supply chain’s performance due to the reason that it
with a low price. would be expensive to shut down a facility or move it to the
The contribution of this paper includes the demand another location. A good location decision will help a
analysis and forecasting to solve the mentioned current supply chain working stably. Their model also help
problem of demand. Moreover, the contribution also identifying plant and warehouse locations between different
consists of the designing the Vietnam milk-apple supply locations which can minimize the total fixed and variable
chain network, which will be beneficial for operating the costs. Furthermore, M. Daskin, L. Snyder, R. Berger (2003)
network of facilities and distribution options. Then, the at the department of Industrial & Management Sciences at
product design and the inventory policy for the proposed University of Northwestern has their work ‘Facility
products of the fruit are also concerned. Location in Supply Chain Design’. The paper has shown
about the importance of facility location in supply chain
2. LITERATURE REVIEW design. They have indicated that a poorly placed plants and
warehouses can result in excessive costs. They began by
With the information from the official website of the reviewing classical models including the traditional fixed
People’s Committee of the Tien Giang province in Vietnam, charge facility location problem and then expanding the
Vietnam milk-apple is long time well-known as the theory of facility location decision to additional features
exclusive product of the Mekong Delta - Vietnam. The fruit such as vehicle routing, inventory management and
is popular and favorable for big fruit, beautiful skin and reliability.
special sweet. For recent years, the fruit is highly evaluated The role of Planning and Forecasting is to forecasting
by not only inland customer but international customers the demand and planning the materials to adapt the
also. However, due to the specification, the fruit cannot be forecasted demand. According to Sunil Chopra and his
stored for a long time more than 2 weeks to serve for partner (2007), demand forecast forms the basis of all
exporting and there is no solution for extending the storing supply chain planning. It means that, the supply chain
time of the fruit up to now. In 2010, the Department of managers need to understand and design with the most
Science and Technology published the report of ‘Extending appropriate forecasting method to manage their supply
and developing Vietnam milk-apple’, the report pointed out chain effectively. Planning – capacity, production,
the importance of storing the fruit and growing the fruit inventory and pricing – is necessary for the supply chain
with the compliance to GlobalGAP and VietGAP. The after demand being forecasted.
report also suggested to establish an official organization Russell and Taylor (2011) have stated “New products
which can function a warehouse to develop post-harvesting and services are the lifeblood of an organization” in their
technology to storing the fruit. In the same year, 2010, with work “Operations Management: Creating Value along the
the collaborative of the Center of Biotechnology and Bio- Supply Chain”. There is a big impact of product design on
engineering, Vietnam South Fruit Institute, Sub-institute of influencing the market of a business and that is the reason
Agriculture and Post-harvesting Technology published a why need to be managed effectively. They have also
handbook of ‘Processing and Storing Vietnam milk-apple’. mentioned about technology decision. Technology is
The handbook has mentioned about the process of another step that influence the cost, speed, quality, and
extending the fruit life and storing condition as well as a flexibility of operations. The capacity of the firm and later
proposed layout of a temporary warehouse which can store interaction was also decided by technology. A technologist
the fruit. However, the fruit is just be able to be kept in or a decision maker have to consider both aspect and
maximum time for 2 weeks and no further research has choose the most appropriate and most efficient choice.
been indicated. In a book of Tompkins and Associates (2008),
Decisions considering the location and capacity “Facilities planning”, facilities planning plays an
allocation of facilities characterize a crucial field of tactical importance role of not only a science but also a strategy. In
planning for supply chain management. An optimization of production, facilities planning related to the defining of
the supply network using a mixed integer linear how the manufacturing facility best supports production.
programming (MILP) model is proposed by Wouda in 2004. They also indicated that facilities planning was a wide
The objective of the study is to evaluating these strategies concept that related to facilities system, facilities layout and
through classification of the optimal number of plants, material handling system. 30 percent of the operation could
locations, and allocations of the product portfolios, while be reduced with a good facilities planning.
minimizing the sum of total cost. Moreover, Sunil Chopra

1130
 Tan and Phong

Year 2012 Year 2013 Year 2014

10000
Quantity (ton)
8000
6000
4000
2000
0
12 1 2 3 4 5 6 12 1 2 3 4 5 6 12 1 2 3 4 5 6
Capacity 1438 5000 4564 5875 2819 2504 0 812 4496 5745 6931 3065 1813 0 857 3697 5135 5090 2570 1668 0
Demand 8062 7865 1383 2328 1602 3168 3936 6627 3858 6179 1776 1404 2679 5496 3832 2691 4872 2921 1943 2394 4184

Capacity Demand
Figure 1: Demand & capacity of Vietnam milk-apple in year 2012, 2013 and 2014.
3. DEMAND FORECASTING & CAPACITY Suppliers
ESTIMATION 1
4 Warehouse
2 3 Hub Candidates
3 Factory Distribution
5000 4085 Candidates Candidates Centers
1
Quantity (ton)

4000 3
3075 2992 3011
3000 1 1 1
4
1625 1800
2000 1130 5 2
1000
6 2 2 2
0
12 1 2 3 4 5 6 7
3
Month 8
Demand 3 3 3
9
Monthly forecasting demand of 4
Vietnam milk-apple in year 2015 10

11
Figure 2: Forecasted demand of Vietnam milk-apple
in year 2015 12
According to the statistics of department of
Figure 3: General framework of the milk-apple supply
Agriculture and Rural Development of Tien Giang province
chain network
- Vietnam, the recorded data of demand and capacity of the
Vietnam milk-apple in year 2012, 2013 and 2014 are shown while the farmer cannot adjust their harvesting time and the
in the figure 1. current fresh milk-apple is just could be stored in maximum
From the graph in figure 1, we can observed that, the 2 weeks.
peak of the demand is often in December (beginning of the
season), February (Lunar New Year month) and June (end 4. MILK-APPLE SUPPLY CHAIN NETWORK
of the season) while the productivity seem to be high in the
duration from January to March. That problem is the root As mentioned, the current supply chain network is
cause which leads to the big variation in the fruit price operating ineffectively. The purpose now is to design a
recent years. supply chain network which find the locations of the
A forecasting using Winter’s model is done to facilities and minimize the facility allocation cost and
estimate the fruit demand in year 2015. With the level, shipment costs subject to a requirement that all customer
trend and seasonality of demand, the Winter model is demands be met.
known to be the most accurate forecasting method with the The supply chain network could be separated in to 5
lowest MAD and the most appropriate method (S. Chopra main units. The units consist of: suppliers unit,
and P. Meindl, 2007). Applying α = 0.05, β = 0.1 and γ = procurement hubs unit, manufacturing plants unit,
0.1, the monthly forecasting result and the total forecasting warehouses unit and distribution centers unit. The functions
demand 17 717 tons of year 2015 is obtained. The goal now of each unit are connected and correlated base on the
is to supply the product as the monthly forecasted demand concept of the supply chain. Figure 3 shows the flow of
shown in figure 2. However, the goal seem to be impossible materials and goods, through a distribution network to

1131
 Tan and Phong

satisfy customer demand. Zp potential capacity of procurement hub at site p

There are 12 suppliers in figure 3 which represent for Ep fixed cost of locating a procurement hub at site p
12 communes grows the milk-apple. The Vinh Kim, Song Ki potential capacity of plant at site i
Thuan, Long Hung, Kim Son, Dong Hoa, Duong Diem, Fi fixed cost of locating a plant at site i
Phu Phong, Bang Long, Binh Trung, Huu Dao, Thanh Phu, We potential capacity of warehouse at site e
Nhi Binh are numbered 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 fe fixed cost of locating a warehouse at site e
respectively. Dj annual demand of distribution center j
Similarly, we have 3 candidates for hub locations, chp cost of shipping one unit from supplier h to
which are situated at the communes with good procurement hub p
infrastructure of transportation. The Vinh Kim, Long Hung, cpi cost of shipping one unit from procurement hub p to
Song Thuan commune are numbered 1, 2, 3 respectively. plant i
Then, we have 3 candidates for plant locations, which cie cost of shipping one unit from plant i to warehouse e
are situated in the local My Tho industrial park. The Vinh cej cost of shipping one unit from warehouse e to
Kim, Kim Son, Song Thuan commune are numbered 1, 2, 3 distribution center j
respectively.
There are also 4 candidates for warehouse location, Decision variables:
which are situated at the communes with good 1 if we locate a procurement hub candidate at site p
infrastructure of transportation. The Long Hung, Song yp 
Thuan, Kim Son and Vinh Kim are numbered 1, 2, 3 and 4 0 otherwise
respectively.
1 if we locate a factory candidate at site i
yi 
Lastly, we have three distribution centers located at
the three biggest cities of Vietnam, which have good 0 otherwise
system of transportation infrastructure, center market
ye 
1 if we locate a warehouse candidate at site e
infrastructure and large demand also:
 Ho Chi Minh city is numbered as 1. 0 otherwise
 Da Nang city is numbered as 2. xhp quantity shipped from supplier h to procurement hub p
 Ha Noi city is numbered as 3. xpi quantity shipped from procurement hub p to plant i
The network conformation problem relates to an xie quantity shipped from plant i to warehouse e
enormous of data, consists of the following input data: xej quantity shipped from warehouse e to distribution
 Locations of suppliers, distribution centers. center j
 Potential locations of procurement hubs, plants,
warehouses and their allocation cost. The facility locating problem can now be formulated
 Transportation costs and frequency of delivery. as this:
Consequently, the model will follows some of the 𝑑 𝑛 𝑡 𝑙 𝑑

assumptions. The assumptions in here include of the 𝑴𝒊𝒏 ∑ 𝐸𝑝 𝑦𝑝 + ∑ 𝐹𝑖 𝑦𝑖 + ∑ 𝑓𝑒 𝑦𝑒 + ∑ ∑ 𝑐ℎ𝑝 𝑥ℎ𝑝

proposed capacity for some of the supply chain units, the 𝑝=1 𝑖=1 𝑒=1 ℎ=1 𝑝=1

secondary data of actual requirements for the proposed 𝑑 𝑛 𝑛 𝑡 𝑡 𝑚

location. With the policy of the local authority in the next + ∑ ∑ 𝑐𝑝𝑖 𝑥𝑝𝑖 + ∑ ∑ 𝑐𝑖𝑒 𝑥𝑖𝑒 + ∑ ∑ 𝑐𝑒𝑗 𝑥𝑒𝑗
ten year, the plant is required to be built with the capacity 𝑝=1 𝑖=1 𝑖=1 𝑒=1 𝑒=1 𝑗=1

of processing an input of 30 thousand tons of milk-apple

per year. The first term of the objective function has the
A linear programming model is built with indices, meaning to minimize sum of the cost of opening a
variables, objective function, constraints with the purpose procurement hub at site p which is decided to be opened or
is to minimize the total cost of the whole supply chain not by the decision variable yp. Similarly, the second term
network (M. Daskin et. al., 2007). has the purpose to minimize sum of the cost of opening a
We have the following sets and indices: plant at site i which is decided to be opened or not by the
l number of suppliers/communes decision variable yi and the third term has the purpose to
d number of potential procurement hub locations minimize sum of the cost of opening a warehouse at site e
n number of potential plant locations which is decided to be opened or not by the decision
t number of potential warehouse locations variable ye. The fourth term minimizes sum of the total cost
m number of distribution centers of transporting a specific quantity of goods from supplier h
Then, we have the following parameters: to procurement hub p. Similarly for the fifth term
Sh supply capacity at supplier h minimizes sum of the total cost of transporting a specific

1132
 Tan and Phong

quantity of goods from procurement hub p to plant i. The Lastly, the constraint (9) means that each of the
sixth term minimizes sum of the total cost of transporting a procurement hub, plant and warehouse is either open or
specific quantity of goods from plant i to warehouse e. closed.
Lastly, the seventh term minimizes sum of the total cost of Table 1 represents the obtained optimal solution of 3
transporting a specific quantity of goods from warehouse e procurement hubs, 1 plant and 1 finished product
to distribution center j. The total cost of the supply chain warehouse.
will be minimized by the objective function subject to Table 1: Obtained optimal solution of unit location.
following constraints:
𝑑 Capacity
Unit Location
Allocation
∑ 𝑥ℎ𝑝 ≤ 𝑆ℎ for h = 1, . . , l (1)
𝑝=1
Vinh Kim commune 10,000
𝑑 𝑛 (10.355111, 106.246814) tons/year
∑ 𝑥ℎ𝑝 − ∑ 𝑥𝑝𝑖 ≥ 0 for p = 1, . . , d (2) Procurement Long Hung commune 10,000
𝑝=1 𝑖=1
Hub (10.363086, 106.261678) tons/year
𝑛 Song Thuan commune 10,000
∑ 𝑥𝑝𝑖 ≤ 𝑍𝑝 𝑦𝑝 for p = 1, . . , d (3) (10.33349, 106.2689) tons/year
𝑖=1 Song Thuan commune 30,000
Plant
𝑛 𝑡 (10.333612, 106.273146) tons/year
∑ 𝑥𝑝𝑖 − ∑ 𝑥𝑖𝑒 ≥ 0 for i = 1, . . , n (4) Finished Product Song Thuan commune 30,000
𝑖=1 𝑒=1 Warehouse (10.333612, 106.273146) tons/year
𝑡

∑ 𝑥𝑖𝑒 ≤ 𝐾𝑖 𝑦𝑖 for i = 1, . . , n (5) 5. MILK-APPLE PRODUCT DESIGN &

𝑒=1 FACILITIES PLANNING
𝑡 𝑚

∑ 𝑥𝑖𝑒 − ∑ 𝑥𝑒𝑗 ≥ 0 for e = 1, . . , t (6) 5.1 Milk-apple Product Design

𝑒=1 𝑗=1
𝑚
According to Dole Food Company, world’s largest
∑ 𝑥𝑒𝑗 ≤ 𝑊𝑒 𝑦𝑒 for e = 1, . . , t (7) producer of fruits and vegetable products, market trend in
𝑗=1
𝑚
recent years has tendency to enjoy freezing and canning
fruit besides the fresh fruit. Moreover, according the Food
∑ 𝑥𝑒𝑗 = 𝐷𝑗 for j = 1, . . , m (8) Manufacturing magazine - one of the large magazine of
𝑗=1
food manufacturing field in the United States, freezing and
𝑦𝑝 , 𝑦𝑖 , 𝑦𝑒 ∈ {0,1}, 𝑥ℎ𝑝 , 𝑥𝑝𝑖 , 𝑥𝑖𝑒 , 𝑥𝑒𝑗 ≥ 0 (9) canning fruit are the most dynamic and largest sectors of
the food industry and the major factor driving the growth of
The constraint (1) stipulates that supplier’s capacity the frozen and canning fruit market is the
must be greater than the total amount shipped by supplier. convenience. Five products are proposed based on the
The constraint (2) states that the quantity shipped from market research of mentioned information source to adapt
supplier to procurement hub must be greater than the various customers’ requirement:
quantity shipped from procurement hub to plant. The  Freezing Milk-apple Pulp in bag.
constraint (3) specifies the quantity shipped from  Fresh Milk-apple in box.
procurement hub to plant cannot exceed the capacity of the  Canning Syrup Milk-apple in can.
procurement hub. The constraint (4) indicates the quantity  Milk-apple Smoothie Powder in bag.
shipped from the procurement hub to plant must be greater  Milk-apple Ice-cream in box.
than the quantity shipped from plant to warehouse. Then, With the development of the proposed products, many
the constraint (5) expresses the quantity shipped from plant of benefits will be obtained. First of all, the biggest
to warehouse cannot exceed the capacity of the plant. The problem – adapting customer demand in year or whenever
constraint (6) points out the quantity shipped from plant to they need – is solved while most of the products has the life
warehouse must be greater than the quantity shipped from span nearly or greater than one year. Then, the
warehouse to the distribution center. The constraint (7) transportation of the fruit become more convenient when
requires the quantity shipped from warehouse to the there is no difficulties in storing and transporting the fruit
distribution center cannot exceed the capacity of the to the customer. The transportation now do not need to
warehouse. The constraint (8) specifies the demand must be depend absolutely on the air. Moreover, the local farmer
covered by the quantity shipped to distribution center. could get great benefit when those fruits which are not

1133
 Tan and Phong

beautiful in skin outside but good of quality inside can be layout is proposed to be 900 m2 in total area after
utilized for processing. The farmers’ fruit selling price is no considering the material flow inside. The group layout is
more affected by the dealers. proposed for the plant by its advantages of combining the
benefit of both the product layout and process layout.
5.2 Facilities Planning The layout of the material warehouse and the finished
product warehouse is also concerned. The operation of the
With the planned product and planned quantity as warehouses, the necessary space for each area are identified
well as the requiring capacity of each unit, the planning for to obtained the best layout.
the facilities of each unit in the designed supply chain
network is concerned.
6. INVENTORY POLICY
We have a total of three procurement hub located at
the three center location of the all growing areas, which is
Inventory management has the function to keep
regarded as the ‘temporary warehouse’ and have the
enough inventory to adapt customer demand as they need.
function to purchase, collect and concentrate the fresh fruits
Changing inventory policies could result in considerable
from local farmers to transport to the plant.
altering the supply chain’s efficiency and responsiveness (S.
The internal are of the hub contains of three main
Chopra and P. Meindl, 2007). This section will consider the
areas: staging area, inspection area and distribution
inventory policy for the five products of the proposed plant
processing area. The staging area has the function as a
as the forecasted demand of year 2015.
buffer area to arrange materials (baskets of fruits) in order
waiting for the inspection. There are 3 lanes with the size
6.1 Choosing inventory replenishment policy
1m width and 5m length. So the total area for the staging
area is suggested for about 15 m2. In the inspection area,
The replenishment policy of the inventory includes
inspectors will check the quality of the materials at the
the determination of reordering time and reordering
inspection area. The total area of the inspection area is
quantity. Such determination will define the cycle inventory,
proposed to be 9 m2. The passed material from inspection
the safety inventory and cycle service level (CSL) also. In
area are put into big basket and waiting for putting on the
this case, there are two main types of policy are considered:
rack to wait for shipping in the distributive processing area.
 Continuous Review Policy: In this type, the
Figure 4 shows the proposed general layout of each hub.
inventory is recorded continuously. The lot size Q is
ordered whenever the on hand inventory decreases to a pre-
Moving Flow of Suppliers
defined level, which is called the re-order point (ROP).
Toilet Moving Flow of Goods  Periodic Review Policy: With this type, the
inventory stock status is checked regularly in periods. After
Distributive Processing Area Inspection Area determining the inventory in stock, the order will be placed
to increase the inventory stock to a specify level.
The Continuous Review policy seem to be applicable
in this case study due to the fact that our products are
perishable item.
FIFO

6.2 Inventory policy for each product

6.2.1 Inventory policy for fresh milk-apple product

Receptionist

Storage Rack Staging Area

Desk

This is the tradition fresh products of all to adapt

ordinary customer demand. The product has short life-time
and follow the demand-chasing policy. It means that
Truck Door whenever the customer needs, the products is supplied as
the required quantity immediately and has no inventory.
We have the planned monthly productivity of the
Figure 4: General layout of each hub.
product is 1500 tons/month.
Based on the forecasted demand result, we got the
demand and supply quantity of fresh milk-apple product in
After defining the manufacturing process and year 2015.
determining the necessary number of machines, the plant Then, the remaining quantity that could not adapt

1134
 Tan and Phong

customer demand is the demand of the remaining products.

Assuming that the demand of the remaining products Thus, the necessary safety inventory to maintain 0.95
follows the planned ratio of the planned production. of CSL is 284 tons.

Table 2: Forecasted demand and planning supply of fresh The re-order point is the level of the inventory at
milk-apple product. which an order is taken to replenish the inventory stock:
ROP = DL + ss = 61 + 284 = 345 tons
Demand Supply
Month
(ton) (ton) Moreover, we have the cycle inventory:
12 4085 1500
𝑄 98
1 3075 1500 Cycle inventory = = = 49 tons
2 2
2 2992 1500
Then, the warehouse required to holds an average
3 1625 1500 inventory of:
4 1130 1130 Average inventory = cycle inventory + safety
5 1800 1500 inventory = 49 + 284 = 333 tons
6 3011 1500
Each of the milk-apple freezing product spends:

6.2.2 Inventory policy for freezing milk-apple 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑖𝑛𝑣𝑒𝑛𝑡𝑜𝑟𝑦 333

Average flow time = = = 0.54
𝑡ℎ𝑟𝑜𝑢𝑔ℎ𝑝𝑢𝑡 618
product and the remaining
month at the warehouse.
The monthly demand of the freezing milk-apple
product is assumed to follow the normal distribution and Then, we have the similar computation for the
have the mean of 618 and the standard deviation of 548. inventory level of the remaining products, which is shown
The inventory follows the continuous review in the appendix of the paper.
replenishment policy. We have the value of the
replenishment lot size:
7. CONCLUDING REMARKS
2𝐷𝑆 2 𝑥 4325 𝑥 30 000
Q= √ =√ = 98 tons
ℎ𝐶 0.2 𝑥 135 000 Vietnam milk-apple is a potential special and the fruit
is now not exploited effectively which resulted in the big
In which, D is the annual demand of the product.
depreciation in the fruit price. With the five proposed
S is the ordering cost.
products from milk-apple in the research, the solution for
h is the holding cost of the product.
the problem seem to be initially found. To obtain the
C is the unit cost of the product.
solution, the research proposed to build the supply chain
network with three procurement hubs; one plant and one
The demand during lead time of the product is
finished-product warehouse in the nearby local area
normally distributed with the mean DL:
industrial park. The plant is suggested with 900 m2 at the
DL = DL = 0.1 x 618 = 61 tons
initial to produce 5 products, which have the capacity of
In which, L is the replenishment lead time.
processing 30 thousand tons of milk-apple per year. A
simulation is also done to check the validity of the
The standard deviation of σL is:
theoretical design.
σL = √𝐿 x σD = √0.1 x 548 = 173 tons
It is suggested to do more study on the topic: the
Desired cycle service level (CSL) of the inventory is
implemented model is the deterministic model with
the probability of adapting order in a replenishment cycle.
uncertainty demand, it is highly recommended to develop
In this case, we require the CSL is 0.95, it means that, 95%
the model to stochastic model with uncertainty demand.
of the replenishment cycle, the warehouse could adapt
Moreover, a better market research is proposed to be done
customer demand and the remaining 5% of being could not
to understand more about the customer demand.
adapt customer demand.
It is highly recommended for the local authority to
Then, with the CSL of 95%, the required safety
consider for implementing the plan. This plan is worked
inventory is:
with the considerable support and expectation from the
ss = Fs-1 (CSL) x σL = NORMSINV(CSL) x σL =
representatives of the milk-apple growing area. If the plan
NORMSINV(0.95) x 173 = 284 tons

1135
 Tan and Phong

is successful, it can be apply for many of the remaining REFERENCES

fruits in the area which is now have the similar problem
with the milk-apple. A. Langevin, D. Riopel. (2005) Logistics System – Design
From the result of this study, we contribute further and Optimization, Springer.
study to the supply chain design of the Vietnam agriculture
products. Our finding also further understanding of the B. W. Taylor, R. S. Russell (2011) Operations
customer demand on the milk-apple and the solution as Management – Creating value along the supply chain,
well, which is critical for the current supply chain. Based John Wiley and Sons, NJ.
on this study, further research to develop the supply chains
for Vietnam milk-apple and other tropical fruits in the south Department of Science and Technology of Tien Giang
of Vietnam could be conducted. province - Vietnam (2010) Extending and developing
Vietnam milk-apple.
APPENDIX
Dole Food Company, Inc., (2011) Annual Report
Table 3: Inventory policy for the canning, smoothie powder http://www.dole.com/
and ice-cream milk-apple products.
Food Manufacturing magazine (2013) Consumer Trends:
Product Milk-apple Growing Demand Boosts Frozen Food Market
Canning Milk-apple
smoothie http://www.foodmanufacturing.com/
milk-apple ice-cream
Component powder
replenishment G. E. P. Box, G. M. Jenkin (1976) Time series analysis:
133 tons 46 tons 53 tons
lot size Q Forecasting and control, revised ed, San Francisco.
demand during
84 tons 14 tons 16 tons
lead time DL J. A. Tompkins, J. A. White, Y. A.Bozer, J. M. A.
standard Tanchoco. (2008) Facilities Planning, John Wiley and
144 tons 29 tons 31 tons
deviation of σL Sons, NJ.
safety
238 tons 47 tons 51 tons M. S. Daskin, L. V. Snyder, R. T. Berger. (2003) Facility
inventory ss
re-order point location in supply chain design.
322 tons 61 tons 67 tons
ROP
Average N. Sinha, J. Sidhu, J. Barta, J. Wu, M. Cano (2012)
304 tons 70 tons 78 tons Handbook of Fruits and Fruit Processing – second edition,
inventory
Average flow Wiley-Blackwell.
0.97 month 0.92 month 1.02 month
time
Official website of Tien Giang province – Vietnam (2014)
http:// tiengiang.gov.vn/

S. Chopra, P. Meindl (2007) Supply Chain Management –

Strategy, Planning and Operation, Prentice-Hall, Upper
Saddle River, NJ.

S. Wichuta, S. Sakesun, P. Pallapat, S. Nikorn (2013) Design

and model of supply chain network for the biodiesel
refinery industry in the southern region of Thailand,
American International journal of contemporary research.

South Fruit Institute of Vietnam (2010) Handbook of

processing and storing Vietnam milk-apple.

1136