Optimized Production
Optimized Production
2-1-2018
Recommended Citation
APA Citation
Youssef, A. (2018).Optimized production and inventory decisions for a mixed make-to-order/make-tostock
ready-made garment manufacturer [Master’s thesis, the American University in Cairo]. AUC Knowledge
Fountain. https://fount.aucegypt.edu/etds/192
MLA Citation
Youssef, Aya M. El-Mehany. Optimized production and inventory decisions for a mixed make-to-order/ make-to-
stock ready-made garment manufacturer. 2018. American University in Cairo, Master's thesis.
AUC Knowledge Fountain. https://fount.aucegypt.edu/etds/192
This Thesis is brought to you for free and open access by AUC Knowledge Fountain. It has been accepted for inclusion in
Theses and Dissertations by an authorized administrator of AUC Knowledge Fountain. For more information, please contact
[email protected].
The American University in Cairo
SCHOOL OF SCIENCES AND ENGINEERING
Winter 2018
Approvals
Thesis Title: Optimized production and inventory decisions for a mixed make-
toorder/make-to-stock ready-made garment manufacturer
Thesis Supervisor:
Name: _________________________________________
Title: _________________________________________
Department: _________________________________________
Institution: _________________________________________
Signature: _______________________Date: _____________
First reader:
Name: _________________________________________
Title: _________________________________________
Department: _________________________________________
Institution: _________________________________________
Signature: ________________________Date: ____________
Second reader:
Name: _________________________________________
Title: _________________________________________
Department: _________________________________________
Institution: _________________________________________
Signature: _________________________Date: ___________
Department Chair:
Name: _________________________________________
Title: _________________________________________
Signature: _________________________Date: ___________
i
ACKNOWLEDGMENTS
Dedicated to my loving and supporting family, wonderful mother, amazing brother
Ahmed and caring uncle Lotfi.
To my father, I wished you were here, I kept the promise and I did it. I am proud that I’m
part of you. My blessed soul father; sheikh Muhammed Muhanna; God gift to me.
My precious friends whom were always beside me, they have always believed in me and
never failed to give me their support; Nesma Abuelkheir, Yosra Zakaria, Hala Salem,
Marwa rashwan, Sara Abdelrazek, Basma Bargal, and Radwa raafat. With my deepest
love and respect, I’d like to acknowledge respectful professors; Dr. Maher Younan, Dr.
Amal Elesawi and Dr. Mustafa Arafa for their generous support and tolerance.
I am particularly grateful for the assistance given by my supportive adviser Dr. Tamer
Abdelmeguid, your support was invaluable.
The gem in the crown Prof. Amr Eltawil, I owe you my deepest gratitude. I can’t find
words to express my gratitude for your meticulous comments, witted advises, continuous
support I’m leaving it to God, rewarding you on his expanse capabilities not my expanse.
II
ABSTRACT
The American University in Cairo
Optimized production and inventory decisions for a mixed make-to-
order/make-to-stock ready-made garment manufacturer Aya M.
Elmehany Youssef
Supervisor: Dr. Tamer F. Abdelmaguid
In this thesis, a production and inventory planning model for mixed make-to-order
(MTO) make-to-stock (MTS) production system in garment industry. Where the
dominant production is typically for the Make-to-order production and the make-to-stock
production is penetrating the mainstream production (MTO) as a way of enhancing the
revenues and maintaining a positive cash flow, that are often degraded due to either
seasonality of demands or production planning challenges. The model considers capacity
planning for the mixed environment when there are predictable fluctuating demands. Due
to the nature of the clothing business, it is challenging for a garment manufacturer to
cope with seasonal changes while having the best capacity utilization.
The literature acknowledges production planning in the garment industry. While a little
focus was for capacity planning for seasonal fluctuating demands. Mathematical
programming for capacity planning in a mixed MTO and MTS garment-manufacturing
environment is a viable approach that can provide effective management decisions that
can help the garment industry to strive in today’s competitive pace.
The proposed model considers distributing the available capacity between MTO and
MTS production and the implications of the costs and revenues for different capacity
distribution. Decisions made on the production amounts, inventory levels and generated
revenues are attained. The model was verified and validated by applying it to a local
ready- made garment factory. The results ensured the validity of the proposed model.
When analysis was made to the parameters that influence the decisions, it was found that
distributing the capacity between MTO and MTS with different percentages had
significant impact on the revenues and costs.
The model was very sensitive to the increases in the fabric price and subcontracting costs
while the overall net profits were not significantly affected by the changes in the
inventory holding cost. Last, this work is useful in helping garment manufacturers adapt
rapidly to seasonal changes by deploying their capacity effectively in favor of their
projected seasonal plans.
III
Table of contents
Approvals .............................................................................................................................
i
IV
3.3 Model formulation .................................................................................................
24
35 Case study
description ............................................................................................... 36
Case 2: No quantity discount for fabric purchased for MTS products. .....................
44
V
Chapter 5 Computational Results and Sensitivity Analysis .............................................
47
5.1 Impact of changing percent occupied by MTO and MTS on the revenues. ............
47
5.2 Impact of distributing capacity between MTO and MTS on the costs. ...................
49
5.3 Impact of changing percent of capacity between MTO and MTS on the inventory
5.5 Impact of increasing the inventory holding cost on the revenues. ..........................
52
References .........................................................................................................................
59
Appendix B: The outputs for multiple of multiple MTO – MTS products ........................
F
Appendix C: Fabric inventory and quantity purchased for the 32 – 64 MTO – MTS ......
12
VI
VII
List of Figures
50 Figure 10: Percentage of capacity distributed between MTO and MTS vs. inventory
revenues ................................................. 52
Figure 12: net profits vs. change in inventory holding cost ..............................................
53
55 List of Tables
VIII
Table 3: Size of the model components ............................................................................
31
Table 13: Base case fabric purchasing amounts and their inventory levels .....................
40
Table 16: Production plan for MTO and MTS produced during period t ....................
41
IX
Table 19: fabric order quantity (FQfkt) solution for 64,128 MTO, MTS .........................
44
Table 20: fabric order quantity (FQfkt) solution for 32, 64 MTO, MTS ...........................
45
Table 21: MTS inventory levels when eqn. (5) is relaxed. ...............................................
46
Table 23: Capacity distributed between MTO, MTS, and their projected revenues ........
47
49 Table 25: Percentage of capacity distributed between MTO and MTS vs. Inv. holding
cost ....................................................................................................................................
50 Table 26: Fabric price percentage increase vs. profits and fabric inventory
cost ............. 51
Table 27: Revenues and costs for inventory holding cost increase ..................................
53
Table 28: Revenues and costs vs. subcontracting cost percentage increase .....................
54
X
X
Chapter 1 Introduction
1.1 Background
The readymade garment industry is considered as a one of the very competitive industries
in the world, Guo, et.al (2006). Its supply chain is dynamic and complex due to the
disparity along the chain. Garment manufacturing is a substantial sector within the textile
supply chain. Apparel industry has two sides of productions. One is the artistic part (high
fashion industry) which is highly perishable and subjected to seasonality. The other is the
Appendix - A
Planning of the ready-made garment production involves many considerations such as
timings of the fabrics supply and designing efficient production schedules that meet
promised delivery dates. The ability to control the inputs and outputs of the apparel
there is variability in the demand types; whether it’s made to order or made to stock; and
amounts. The garment manufacturing is considered labor intensive due to the large
Garments produced to fashion are subjected to high variety of taste, which implies more
frequent garments to be available in stock prior to season and preferably an end of season
accelerated market.
Often there are products produced to order and garments produced to stock. An effective
factory is very crucial. Production managers are striving to meet orders due dates and at
the same time achieve their production plan targets for selected seasonal products, that
Production of garment undergoes a series of processes. Figure 2 shows that the input to
the process is the fabric and the output is the desired garment product. In many cases both
types of production systems are tied up to the same process. The typical bottleneck in a
garment production process is the sewing stage/ stitching. As it is considered the longest
processing stage among garment-manufacturing stages, and that is why the industry is
characterized by being labor intensive. Shorter time on the sewing requires highly skilled
workers and therefore very high costs. It also needs attention to the quality of the stitched
piece of garment before proceeding to the next operational stage, due to the difficulty of
repairing any uncovered defects after an item being sewed. However, workers operating
at the cutting stage are the most expensive. Typically, the minimum batch size of a
production
cutting spreading marker making
pattern
button and
packing / final button
dispatching inspection finishing
hole
Delivery
/warehouse
2
Figure 2: Garment production processes
• Design/ Sketch
The design or the sketch of the garment is the primary step in garment manufacturing.
• Pattern Design
Each component of the garment has its own pattern. That pattern is sketched on a hard
paper with the exact dimensions. Seam allowance, trimming allowance, ease allowance,
pleats and any special designs are all included in the pattern dimensions.
• Sample Making
The pattern is placed over the fabric, which is cut into the pattern form, and then the
• Production Pattern
After the customer approves the sample pattern, production of the pattern starts.
Sometimes different patterns that include any modifications required or suggested by the
• Marker Making
All required sizes of the same model and same fabric are detailed into its pattern design.
Pattern details mean arranging pieces of the same garment in a way so as to minimize the
3
• Spreading
It is the process prior to fabric cutting directly where the fabrics desired to be cut is
placed over each other in layers. The height of all the whole layers should not exceed the
• Fabric Cutting
The marker paper of the desired pattern is pinned over the fabric layer as accurately as
possible. The straight knife cutting scissor is then used to cut out the fabric into the
required garment pieces. This part of the process needs highly skilled labors for its
accuracy.
• Sorting/ Bundling
Every component of the garment is sorted together in stack form according to their color
Sewing is the process of assembling garment components using different types of sewing
• Inspection
At the inspection point, the sewed garments are carefully inspected to identify defects that
may have occurred in preceding stages. Some defects such as wrong measurements can
• Pressing
4
In this stage, the pressing machine is used if pressing is needed for specific garments like
• Finishing
inspected to make sure there are no defects. Then it is clean then it is ironed and folded to
• Final Inspection
It includes checking the fine garment details such as buttons, zippers, labeling, ironing
and cleaning.
Products are packed after the final inspection. Packing may be sorted according to
colors or size and then bundled in to boxes. The packaged garments are either
capacity planning in such an environment is very challenging particularly for stages with
manual operations. Manufacturers are always trying to meet new trends and therefore the
There are important decisions that have to be taken cautiously as they have direct impact
on the costs and revenues. Decisions like determining optimal production times and
quantities to meet due dates for MTO products. Moreover, having an updated seasonal
plan for new seasonal products so as to preserve a competitive edge while at the same
5
1.2 Problem description
Egyptian manufacturers were greatly affected by the economical stagnation the past few
years, as shown in the data in figure 3 according to the central bank of Egypt. Factories
that depend solely on the local market have been greatly affected from the egyptian
pound flotation, for fabric prices have notably increased. Fabrics required for large
orders, can be supplied on installments along the production plan of the order. However,
consecutive rises in the fabric prices encounter remarkable drawbacks in the cash flow
that disables a smooth production to meet up with the upstream orders. The shorter the
planning, the less risky the management will be subjected to the fluctuating fabric prices.
6
This work is motivated by the drawbacks in the net profits experienced by some of the
ready-made garment factories. Factories that were affected by several and sequential
financial strikes, resulting from continuous payment delays along with pressuring
The main challenge would be in the complexity of managing both production types
within a short planning horizon with limited resources. Challenges like meeting orders
due date, ability to produce forecasted demand within the season and succeed in
growing size of the population is an excellent asset that entails wide and varying
marketing chances. Figure 4 shows the increasing trend of the population which
7
Typically, ready-made garment industry in Egypt is managed by experience. Relying its
matter that allows it to cope with global developments in the field. Thereby, the
beginning of the plan implementation was initiated from a deterministic model just to
have a hand on the most frequent problem to tackle them together and then there would
be further improvements that would consider the uncertainty of different aspects. At that
The rest of this thesis is organized as follows. Chapter 2 is a review of the literature
addressing the thesis topic. Chapter 3 is a presentation and explanation of the developed
model. The model validation and verification is highlighted in chapter 4 followed by the
computational results and analysis in chapter 5. Closing with the conclusions obtained
from the work and further recommendations for future work in chapter 6.
Three appendices are provided at the end of the thesis that present the input data and
8
Chapter 2 Literature Review
Make to order production does not always guarantee an on spot payment. It just ensures
that there will be a cash inflow during a future period, which cannot be the only reliable
source of revenue for sustaining a garment manufacturing facility. Therefore, firms plan
to produce for stock. Although there are trade-offs for both modes of production,
planning a made to order or made to stock only manufacturing environment is easier than
planning both types of production in the same environment. This is very clear throughout
the relevant literature. Research in the production-planning field for either MTO or MTS
production systems is addressed quite often in literature than that with both modes of
production within the same system. Limited number of papers that do consider them
This chapter introduces a review of the aforementioned topic. It is divided into three
primary areas related to the research at hand: the first deals with production planning in
the garment industry, the second part focuses on production planning related to MTO
environments and the third part deals with garment production planning for MTO and
Several research papers covered in production planning in the apparel industry. Wong and
Chan (2001) proposed a heuristic framework that included master production scheduling
(MPS) for an apparel factory. Its target was to minimize the total cost whether the
demands were completed before or after their due dates and to minimize the presewing
9
Mok (2011) stated that the apparel manufacturing environment is a fuzzy environment
and hence an effective production planning for such an environment would be difficult.
manufacturing environment where he applied fuzzy set theory, genetic algorithms (GA)
Guo, et.al (2006) used a job shop scheduling (JSS) model for a mixed- and multi-product
assembly apparel environment to meet demands. The proposed model was solved by
applying GA. The objective of the proposed model is to minimize the total penalties of
earliness and tardiness by deciding orders’ production starting time and the way of
Chen, et.al. (2012) developed a grouping genetic algorithm (GGA) for assembly line
balancing problem (ALBP) particularly for sewing operations in the garment industry.
Mok, et.al. (2013) used also GA in conjunction with group technology to obtain
automatic job allocations that ensures an effective utilization of resources and jobs
completion.
Ünal, Tunali and Güners (2009) research also suggested a heuristic algorithm for the line
balancing problem in the apparel industry and tested its validity by using various line
Choy, et.al. (2011) combined genetic algorithm with an optimization model in a hybrid
scheduling decision support model (SDSM) that solved the scheduling problem in a made
to order work place. The model considered job tardiness and process variations to achieve
10
Often the production planning reported in literature is devoted to make-to-stock (MTS)
systems differ from that of MTS. In MTO, delivery dates are crucial, although the
demands are uncertain in quantities and times; hence, it requires smart detailed
production plans that enable full control of the production inflow and outflow.
Yeh, (2000) presented an approach for planning multi-product levels that uses a bill of
material and data routing in organizing production data to accommodate customer orders
specifications. This in turn enables them to negotiate order adjustments that might include
allowing for changes in delivery schedules. Due to different constraints, the system is
able to fulfill only some of them and the rest are rejected. Under this condition, MTO
companies have to accept an optimal combination of the arriving orders so that their
profit and share in the competitive market increase. The main criteria to select this
optimal combination in such environments are short delivery times and high quality.
Therefore, companies that produce according to MTO systems, are not able to predict the
arrival time of orders and have to deliver the arriving orders quickly. Such companies
need a decision-making structure which helps them manage the arriving orders to meet
Ebadian, et.al (2008) introduced a decision making structure for the order entry stage for
a MTO system. The decisions were based on two factors. One was the price; the other
was the delivery time whether fixed or flexible (if it could be negotiated). The
mathematical model adopted was mixed integer programming. The structure made the
decision on the arriving orders based on certain factors; determining the rejected
11
undesired orders and computing the prices and delivery times of the accepted orders
bounded by some constraints. The main challenge of such a MTO program was to
process directly the arriving orders that are only profitable and feasible to the system.
therefore a decision making structure to improve the decision quality at the order entry
manage arriving orders but they divided the structure into two main phases. The first
phase was designed to arrange orders according to order times, material arrival, and due
dates. The second phase was decided upon orders using a hybrid methodology that
combines analytical hierarchy process (AHP) with technique for order performance by
Hemmati, Ebadian and Nahvi (2012) also created a decision making model to manage
arriving orders, but they first divided them according to order priority. High or low
priority depends on the order characteristics and TOPSIS. Orders were divided according
to their due dates, material requirements and availability. Then a rough cut capacity plan
for the orders was used to check which order would be feasible for the system.
Ebadian, Rabbani, Jolai and torabi (2009) proposed a hierarchical production planning
structure in MTO companies that consisted of three decision levels: order entry, order
release and order sequencing. Their target was to meet due dates by improving delivery
date performance using a smooth production schedule. The last two levels were validated
through numerical trials and simulation. The aim of the proposed hierarchy is to manage
12
arriving orders using certain decisions per level, so as to have shorter, reliable delivery
dates.
Soman, Donk and Gaalman (2004) developed a hierarchical planning system to decide
which products to make- to- stock and which to make- to- order in the food industry. The
authors considered capacity restrictions and varying set-up time between different
products where the decision plan related to food. In the same fashion, Rajagopalan,
(2002) developed a nonlinear integer programming model to decide which orders to make
to stock and which to make to order, their inventory levels and selecting a production
policy for the made to stock items. Rajagopalan (2002) put a penalty constraint to the lead
time for MTO items and stated that he could have made MTO demands function in the
lead time of each item in the demand. One of the aspects he considered in his model was
the fractional lot sizes, re-order points and continuous distribution which forced model
approximations. However, various aspects of the model presented here, such as the cost
expression, fractional lot sizes and reorder points, and continuous-demand distributions,
Wang and Rosenshine (1983) selected a heuristic rule for given orders, where some are
with due dates (made- to -order) and others are not (made to stock). Their objective was
to schedule the orders so that the mean flow time was minimal while satisfying their due
dates. There were two types of conflicts that appeared to them, the first was related to the
Kaminsky and Kaya (2009) provided a decision guidance on when to use made to order
13
analyzed the benefit of using each system depending on specific circumstances, and how
to operate the overall system in order to minimize the costs. Kamisky and Kaya also
quoted due date per customer at arrival time. All that was explored analytically,
inventory, scheduling, lead time decisions but in the context of supply chains where they
Rafiei, Rabbani and Kokabi (2014) addressed production planning of a multi-site hybrid
Carr and Duenyas (2000) addressed the problem of modeling complex admission control
and sequencing of different product segments. This was done by developing a M/M/1
queueing model to help the firm take former decisions into account when deciding which
type of product to produce next and the annual quantities for MTS products. One of the
major gaps which should be taken into consideration was the setup cost for different
products. This would influence the production sequence and how decisions were made to
Zhang, Zheng, Fang and Zhang (2015) proposed a mixed integer non-linear programming
model aimed at solving the multi-level inventory matching problem. They focused on
order planning for a hybrid MTO and MTS production planning strategies, as applied to
the steel production environment. Multiple objectives were considered, such as penalty
cost of earliness and tardiness, production costs, inventory matching cost, and order
cancelation penalty. The infeasibility and inventory re-matching were treated through a
14
Beemsterboer, Land and Teunter (2016) examined the advantages of hybrid planning
approaches for both MTO and MTS by developing a Markov Decision process model.
The system determines when to manufacture MTO and when to produce MTS products,
Morikawa, Takahashi and Hirotani (2014) investigated different policies for MTS
Their primary objective was to minimize the average orders' lateness with less MTS
inventory. The performances of these policies were tested using simulation experiments.
The semi-finished products were considered the MTS items and restrictions applied
according to specifications and amounts. The criteria for selection were based on different
rules”.
Zhang, et.al. (2013) analyzed the performance measures of a hybrid MTS-MTO system
for the same product using an analytical model. The dynamic switch between MTO-MTS
Günalay (2011) used two scheduling strategies (first in first out and cyclic service) to
decide which scheduling policy to use for MTO versus MTS products in a single facility.
The decision for both policies was based on the total cost, inventory and order delay cost.
Rafiei and Rabbani (2011) developed a mixed integer linear programming model to
decide order partitioning, and to determine order penetration point location for a hybrid
combined MTO and MTS but focused on the food industry. In 2007 they tackled the
15
combined MTO/MTS situations by identifying possible analytical decisions for shortterm
batch-scheduling using heuristics. The updated approach was adopted when the
researchers found the hierarchical frame work suggested by them in (2004) was generic
Ohta, Hirota, & Rahim, (2007) analyzed a multi-production inventory policy for MTO
versus MTS. The analysis was based on a queuing model where the optimality condition
was based on which product is made-to-order and which is to stock. They computed the
approach to decide which item will be made- to-order and which to stock using a strategic
method strengths and weaknesses, and the opportunities and threats (SWOT) analysis and
decision for orders partitioning by producing quantitative values for the SWOT factors.
However the novel approach did not consider important constraints such as the firm’s
Hadj, Delft and Dallery (2004) used two different scheduling rules for combined MTO
and MTS manufacturing system. MTS policy went under FIFO scheduling whilst MTO
was considered of low volume and went under priority rule policy. Optimum solutions for
Van Donk (2001) indicated that managers in the food industry found difficulty in
deciding which products to make to order and which to stock. Van Donk developed a
framework that helped make such decisions. Taking into consideration the market
characteristics and the production process. The developed framework was based on the
16
concept of the general decoupling point which was adapted to the specifications of the
food industry.
Arreola-risa (1998) also used an analytical way to study the optimality conditions for
MTO versus MTS for multiple products. The production times were general random
variables, and the demands have different arrival rates and are independent Poisson
processes. Khakdaman, et.al. (2015) developed a robust model for hybrid MTO/MTS
multi-product firm. They incorporated suppliers, processes and customers in the presented
applying it to an industrial case. Eltawil and K.W. (2011) proposed a hierarchal frame
work for production planning for a hybrid MTO and MTS production in the textile
In the apparel industry, most of the research has focused on applying heuristic and
hierarchy techniques to help the decision maker achieve the production goals. It is
apparent that production planning in a combined MTO and MTS garment manufacturing
environment was not tackled enough in literature. Considering capacity utilization with
financial aspects were not jointly addressed. Thus targeting certain or near optimal
17
production and capacity plan for a mix of MTO and MTS demands, is still an open
question.
The objective of this work is to help the management team in a garment factory have
practical insights on the decisions to take based on the costs versus the expected revenues
for arriving orders (MTO) versus planned production (MTS) for a season.
The presented model in this study is devoted to medium size factories which have limited
resources, and cannot afford more expenses to improve their productivity. At the same
time, they cannot adopt sophisticated techniques used by large companies nor can they
deal easily with complicated heuristics. Yet, they still need to maximize their revenues as
far as possible, while considering their capacity constraints and meeting their production
targets. The unique problem facing such garment factories is not well addressed in the
literature, which is production and capacity planning for a mix of MTO and MTS
production while considering influential parameters that governs the garment production.
This research is mainly concerned with the decisions taken at the aggregate planning
(MILP) model that is formulated to take into account the important aspects that influences
the decisions at that level of planning and provides an optimum production plan for the
addressed problem.
18
Management of MTO and MTS production together in a readymade garment factory is a
challenging problem. Particularly if the planning horizon is short, due to the nature of the
industry, and if one of the systems was the only applicable kind of production. Various
garments require different number of labor hours and amount of fabric per garment and
the factory has limited capacity and financial resources finances. Also, once fabric is
ordered, the factory must provide the enough cash to pay for it.
The fabric is acquired from two main sources; a retailer or a wholesaler. The retailer price
is 10-15% above the wholesale price. In order to purchase from a wholesaler the amount
purchased should not be less than 30 meters of fabric of the same type.
The amount of products decided for the MTS production, determines whether its fabrics
will be purchased from a retailer or a wholesaler. Past data showed that the MTS fabrics
were bought from a retailer without any discounts, due to the small quantities and in other
season was due to the occurrence of a variety of MTS products with limited amounts per
product.
Fabrics for MTO garments are purchased only if the order is received, while for the MTS
case, fabrics are purchased based on the cash availability and production capacity.
Therefore, it is essential to identify the time, and amount of fabric purchased while
observing the cash availability. Smooth production requires that fabric supply is
The management team aims at having a production plan for the factory that considers a
successful operation of seasonal MTO and MTS production simultaneously, within the
19
available resources. To achieve an optimum plan, major trade-offs from the interaction of
both production types and their implications on the cash and revenue at the end of the
season have to be considered. Trade-offs such as, capacity allocation for both production
types during different seasons, producing an order in an early period and storing it or
Moreover, what if the forecasts for MTS are high and the MTO demands are low, would
the cash available at the beginning of the season and the cash generated from the MTO
along the planning horizon be sufficient to produce targeted stocks and still generate a
profit?
Since fabric comprises a high percent of a garment cost, the amount of products decided for
MTS production determines whether its fabrics will be purchased from a retailer or a
wholesaler.
Fabrics for MTO garments are purchased only if the order is received, while for the MTS
case, fabrics are purchased based on the cash availability and production capacity.
Therefore, it is essential to identify the time, and amount of fabric to order while
Planning the production and delivery of garments under the fore-mentioned trade-offs,
while considering the cash flow during the planning horizon was a challenging problem.
Therefore, decisions need to be made to deal with these trade-offs. Such as the amount of
fabric supplied and its ordering time, inventory levels (fabrics inventory, the work in
20
quantities, periodical cash availability and the optimum amount of MTS production along
The objective of the developed model is to maximize the net profits for various garments
required for either MTO/MTS customer, while maintaining a positive cash flow
The proposed model is a deterministic model that is developed for a mix of MTO and
MTS production within the limited resources. The model maximizes the net revenues
resulting from the MTO and MTS sales along the planning horizon. It also provides an
The main difference between MTO and MTS in the model is that the MTS products are
produced and stocked along the planning horizon to meet the forecasted amounts, and no
21
8. Subcontracting is allowed for MTO products only.
13. Safety stock is not considered for neither MTO fabrics nor MTS products.
Set index
𝑰𝑪𝑶𝒎𝒕 Inventory holding cost per unit of MTO product m during EGP/unit/week
period t.
𝑰𝑪𝑺𝒋𝒕 Inventory holding cost per unit of MTS product during EGP/unit/week
period t.
22
𝑹𝑪𝑶𝒎 Regular time production cost per unit of MTO product m. EGP/unit
𝑹𝑪𝑺𝒋 Regular time production cost per unit of MTS product j. EGP/unit
𝑶𝑪𝑶𝒎 Over- time production cost per unit of MTO product m. EGP/unit
𝑶𝑪𝑺𝒋 Over- time production cost per unit of MTS product j. EGP/unit
𝑺𝒃𝑪𝒎𝒕 Subcontracting cost for MTO product i during time period EGP/unit
t.
𝜶𝑶𝒇𝒎 Amount of fabric f used to make one unit of MTO product m2.fabric
m.
wf Warehouse space needed per square meter of fabric f m2/ m2of fabric
𝒗𝑺𝒋 Storage space requirements per unit of finished MTS product m2/unit
j
Vmax Maximum storage capacity for MTO and MTS final products m2
23
rfk Purchase price r of fabric f at level k, where k=1,2, indicating EGP/ m2
the two pricing levels.
𝑰𝑶𝒎𝒕 WIP Inventory level of MTO product m by the end of period t unit
𝑰 𝑺
𝒋𝒕 WIP Inventory level of MTS product j by the end of period t unit
Regular time production quantity of MTO product m during period t unit
24
3.3.4 The objective function
The objective function aims at maximizing the firm’s total profits ‘P’ which is the net
value achieved from subtracting potential cost elements from sales revenues.
The Total Revenues = the total sales value at the end of the planning horizon for made to
The total production costs are the sum of the regular costs, overtime costs, fabric costs,
inventory holding cost for fabrics, work in process and final products over the planning
− ∑𝑗∈𝐽 𝑅𝑗𝐶𝑆 ∑𝑡𝜖𝑇 𝑅𝑗𝑡𝑆 − ∑𝑗∈𝐽 𝑂𝑗𝐶𝑆 ∑𝑡𝜖𝑇 𝑂𝑗𝑡𝑆 − ∑𝑡∈𝑇 ∑𝑓∈𝐹 𝑟𝑓𝑘 𝐹𝑄𝑓𝑘𝑡
3.3.5 Constraints
Initial fabric inventory is indicated by constraint (1) while constraint (2) represents the
material balance constraints for MTO and MTS products. Initial inventory for MTS
25
equation for meeting MTS forecast. Constraint (5) indicates MTO demand satisfaction
constraint.
Equations (6) and (7) are for the capacity constraints for regular and overtime products
respectively. Fabrics storage capacity constraint is denoted by equation (8). The storage
capacity for MTO and MTS final products is illustrated by equation (9). Equations (10)
and (11) are developed for the quantity discount on fabric purchase. where k represents
the two price levels, k =1 means that no discount is offered for a quantity less than qf , as
illustrated by equation (10), while k=2 means that the amount purchased is greater than qf
and therefore the discount is offered, equation (11). Equation (12) represents the initial
The cash balance for the first four periods of the planning horizon is presented by
equation (13). Equation (14) indicates the cash balance from period 5 to the end of the
planning horizon, where the MTS sales take place with the MTO sales. The final cash at
the end of the planning horizon should be greater than or equal an amount CT, as denoted
by equation (15).
Equations (16) and (17) satisfy the minimum batch production for MTO production. MTS
minimum batch production is presented by equations (18) and (19) is for non-negativity
constraints.
26
− ∑ ∑(𝑹𝑪𝑺𝒋 𝑹𝑺𝒋𝒕 + 𝑶𝑪𝑺𝒋 𝑶𝑺𝒋𝒕 ) = 𝑪𝑯𝒕 , 𝒇𝒐𝒓𝒂𝒍𝒍 𝒕 𝝐𝒕𝟏 , 𝒘𝒉𝒆𝒓𝒆 𝒕𝟏
𝒋𝝐𝑱
= { 𝟏, 𝟐, 𝟑, 𝟒} (𝟏𝟑)
27
= 𝟓, 𝟔 … . 𝟏𝟐. (𝟏𝟒)
𝑭𝑸𝒇𝒌𝒕, 𝑰𝑭𝒇𝒕, 𝑪𝑯𝒕 , 𝑰𝑶𝒎𝒕 , 𝑰𝑺𝒋𝒕 , 𝑹𝑶𝒎𝒕 , 𝑹𝑺𝒋𝒕, 𝑶𝒎𝒕𝑶 , 𝑶𝑺𝒋𝒕, 𝑺𝒎𝒕 ≥ 𝟎 (𝟐𝟎)
3.4 Model Characterization
The size of the MILP model developed in the previous section can be determined from
the cardinality of the sets used in the model as detailed in table 3. Likewise, the total
number of integer variables, binary variables in addition to the number of constraints can
Variables
FQfkt |F||K||T|
IFft |F||T|
𝐼𝑚𝑡𝑂 |M||T|
𝐼𝑗𝑡𝑆 |J||T|
28
𝑅𝑚𝑡𝑂 |M||T|
𝑅𝑗𝑡𝑆 |J||T|
𝑂𝑚𝑡𝑂 |I||T|
𝑂𝑗𝑡𝑆 |J||T|
Smt |M||T|
CHt |T|
Constraints
29
Minimum batch constraint for MTS |J||T|
Given that for the current model, the planning horizon T is 12 periods, the number of
fabric types F used is only four types and the price levels k are two levels, then the size
of the model is controlled by the number of MTO and MTS products. Accordingly, for
different number of MTO and MTS products, the problem size can be determined before
hand as laid out in table 5, and the time taken by CPLEX to converge to an optimal
integer solution.
The problem with different sizes was solved on a server with the following specifications:
30
AMD opteron ™ processor 6174 2.2GHz (2 processors), 128 GB (RAM) - system type:
Since the model size can be anticipated from the cardinality of the variables and
constraints, figure 2 depicts the pattern of the problem size when the number of products
45000
40000
35000
30000
The Model was solved using the optimization solver IBM® ILOG® CPLEX®
Optimization Programming Language (OPL) built-in tools for tuning and conflict
detection enabled a rapid and accurate deploying of the model for its high performance
solver. CPLEX can choose the best algorithm from the multiple algorithms provided for
various kinds of models [1]. Computational experiments for different number of MTO
31
and MTS products were examined. Inputs and Outputs for multiple of products are given
in Appendix A and B.
Chapter 4
This chapter describes the computational working plan in order to verify and validate the
proposed mathematical model. The first section of the chapter involves the verification
phase where a base case was examined in order to verify the model and ensure the
correctness of the solver output. The second section of the chapter is concerned with the
validation stage for the model and it included three cases. The first case is considered to
be the essential case as it used actual data and its output was compared with the output
from a real life case. The second case was implemented to make sure that the quantity
discount constraint is working correctly and the third case was carried out to examine the
An important aspect of the verification procedure was to confirm that the model was
encoded properly before applying different cases. Using CPLEX optimization studio
V12.7.1, exporting the model to an LP file is a way to log exactly how it is read and
32
The model was tested using the data for an industrial case, which is treated as the base
case for the computational experiments. The plant and input data are presented in the
following section followed by the results and discussion of the base case output.
manufactures different types of garments. The factory used to produce only make-toorder
production for the last five decades. Although make-to-order production guarantee its
market, but it often encountered stressing operational schedules so as to meet the due
dates. In the recent years, the factory faced many problems due to either inability to meet
due dates or difficulty in receiving orders’ cash promptly after its delivery. Consequently,
many obstacles were raised due to the unavailability of sufficient cash for future
scheduled orders. Therefore, the number of received orders was diminished, which
team sought of establishing a brand for children wear, which will be the MTS production
stream, unlike the MTO products, which are service wear. The start-up cash for the
maketo-stock production was relevantly small for the new production stream. Make-to-
stock production type differed than those for make-to-order, the make-to-order was
service wear and required experienced workers for its complicated technical operations,
while the make-to stock is often children wear and was technically much simpler. The
attempt was taken by trying only two products (MTS) per season. From the historical data
for the past two years, the sales made by the selected MTS products; although the
production was not as big as MTO; have built rapid cash inflow within its season. In
addition it has received remarkable high demands, hence, the planner need to have
33
smarter decisions for managing the MTO and MTS together within the available capacity
firm.
The proposed model was applied to AMDC Company; it is one of the leading companies
company was established in the early 1980s producing different service garments to
different governmental sectors. All of the company production until 2014 was only based
on MTO strategy, and recently they considered producing to stock as part of their
production plan in order to enhance their revenues. Since the facility included the mixture
of both production types therefore it was a suitable application for the proposed model.
Based on historical demands for MTO, there were two main products, which were highly
demanded; which are the suits (MTOprod1) and the two-piece overalls (MTOprod2). In
the computational work, those were the products considered for MTO along with the
seasonal products for MTS. The production capacity data obtained from the factory were
as shown in table 6 for the regular, overtime and storage capacities. The regular hours are
48 hours per week and the over-time hours are 12 hours per week.
34
In all the computational runs, a period is one week, and there are four common types of
fabrics that are used for MTO or MTS production. Therefore, those are the only ones
considered. Fabric input parameters are given in table 7 Followed by MTO and MTS
35
36
Table 10: Base case optimal results
products variables amount period
𝑹𝑶𝟏𝒕
1000, 215, 806, 1000, 99, 1000, 1000 1, 2, 3, 4, 6, 8, 12
𝑶𝑶𝟏𝒕
0
MTO 1
𝑰𝑶𝟏𝒕
1000, 695, 1501, 1, 1, 100, 100 1, 2, 3, 4, 5, 6, 7
S1t
0
𝑹𝑶𝟐𝒕
785, 206, 401, 1000, 500, 500, 1000 2, 5, 6, 7, 9, 10, 11
𝑶𝑶𝟐𝒕
0
MTO 2
785, 270, 476, 877, 500, 1000, 2000 2, 3-4, 5, 6, 9, 10, 11
S2t
723 7
𝑹𝑺𝟏𝒕 400, 500, 500 5, 6, 10
MTS 1 𝑶𝑺𝟏𝒕 0
𝑰𝑺𝟏𝒕 250
𝑹𝑺𝟐𝒕 1
37
The input data distribution was based on historical demand patterns while for the MTS
was mainly desired to be produced in the first quarter of the planning horizon. The results
for the base case are presented in table 12 with an optimal integer objective of 153226
EGP. Where the model decisions to produce in regular, overtime /subcontract or hold in
Results for amounts of fabric purchased per period and their inventory levels are
Table 11: Base case fabric purchasing amounts and their inventory levels
FQf2t
Periods
Fabric 1 2 3 4 5 6 7 8 9 10 11 12
F1 4159 0 0 0 0 17 3000 0 1500 1500 3000 0
F2 9360 0 0 0 0 0 0 3000 0 0 0 3000
F3 1800 0 0 0 0 0 0 0 0 1000 0 0
F4 1313 0 0 0 650 217 258 275 0 0 0 0
Periods
Fabric 1 2 3 4 5 6 7 8 9 10 11 12
F1 4159 1804 1804 1804 1186 0 0 0 0 0 0 0
F2 6360 5715 3297 297 297 0 0 0 0 0 0 0
F3 1800 1800 1800 1800 1000 0 0 0 0 0 0 0
F4 1048 748 282 0 0 0 0 0 0 0 0 0
The cash flow per period for the base case is indicated in figure 3
38
cash/period
500000
400000
300000
200000 cash/period
100000
0
1 2 3 4 5 6 7 8 9 10 11 12
Table 13: 𝐏𝐫𝐨𝐝𝐮𝐜𝐭𝐢𝐨𝐧 𝐩𝐥𝐚𝐧 for MTO and MTS produced during period t
39
Model validation was performed in order to validate the solver output solutions and
evaluate expected outcomes from real and extreme scenarios that help at highlighting the
limitations of the underlying model. The test cases indicated in table 17 were used for
that purpose. They impart the objectives mentioned in the second column of the table and
1. Limited MTO and The objective of the case is Zeroes in most periods and
considerable space to show that depending small revenue is realized.
capacity. solely on the MTO; Actual data was used for
especially when the arriving this case.
orders diminish; was in
efficient.
2. No quantity discount To ensure that the fabric Fabrics required for MTS
for fabric purchased quantity discount equations products will be purchased
for MTS products. are working correctly. at price level rf1 and the
amounts will be decided for
their perspective periods for
the decision variable FQf1t .
3. Case where the To show how the MTS is Higher inventory costs and
model chooses profitable provided that lower revenues.
mainly there is enough money, that
MTO/MTS. can funds its production and
its market is guaranteed.
The first case in the validation is the case when the factory used to produce only MTO
products. The table below represents actual demands from historical data and their output
solution.
40
Table 15: Validation - Case 1
MTO 1 0 0 0 80 0 0 0 0 0 666 0 0
Demand
MTO 2 0 0 0 0 0 0 0 0 0 0 800 0
products variables 1 2 3 4 5 6 7 8 9 10 11 12
𝑹𝑶𝟏𝒕 746 0 0 0 0 0 0 0 0 0 0 0
𝑶𝑶𝟏𝒕 0 0 0 0 0 0 0 0 0 0 0 0
MTO 1
𝑰𝑶𝟏𝒕 746 746 746 666 666 666 666 666 666 0 0 0
S1t 0 0 0 0 0 0 0 0 0 0 0 0
𝑹𝑶𝟐𝒕 254 196 0 0 0 0 0 0 0 0 137 0
𝑶𝑶𝟐𝒕 0 0 0 0 0 0 0 0 0 0 213 0
MTO 2
𝑰𝑶𝟐𝒕 254 450 450 450 450 450 450 450 450 450 0 0
S2t 0 0 0 0 0 0 0 0 0 0 0 0
Results displayed for case 1; table 18; confirmed the results from the factory that there
was an obvious large capacity not in-use as indicated by the zeroes for many periods and
although the model solution moves towards maximizing the revenues, there were barely
any revenues realized which support the attempt of initiating a MTS production stream.
Fabric quantity discount constrained was confirmed as shown in the solution for the 64,
128 MTO, MTS products and the 32, 64 as well, as tabulated in table 19 and table 20
respectively. The decision for purchasing fabric quantity at different price levels is binded
to the cash availability, fabric storage capacity and demands due dates.
Table 16: fabric order quantity (FQfkt) solution for 64,128 MTO, MTS
FQf1t
1 2 3 4 5 6 7 8 9 10 11 12
0 0 0 0 0 0 0 30 30 0 0 0
41
0 40 40 14 0 40 0 40 0 0 40 0
0 0 0 0 0 0 0 0 100 0 0 100
FQf2t
1 2 3 4 5 6 7 8 9 10 11 12
509756 0 0 0 0 0 0 0 0 0 0 0
75208 0 0 0 0 0 0 0 0 0 0 226
Table 17: fabric order quantity (FQfkt) solution for 32, 64 MTO, MTS
FQf1t
1 2 3 4 5 6 7 8 9 10 11 12
0 0 0 0 0 0 0 0 0 30 0 0
0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 2 0 100 100 0 0 0
0 0 0 0 0 100 0 0 0 0 0 100
FQf2t
1 2 3 4 5 6 7 8 9 10 11 12
296956 0 0 0 0 0 0 0 0 1260 0 0
46450 0 0 0 0 0 0 0 0 0 0 100
42
Case 3: Case where the model chooses mainly MTO/MTS.
In order to enable the model to choose whether to produce MTO or MTS, constraint
number (5) was relaxed, thereby MTS was produced in the remaining capacity resulting
in high inventory levels as shown in table 10, which impacted the revenues; 1,345 EGP.
That is due to the limited initial cash and capacity constraints. Table 21, shows the
increasing inventory levels for the four MTS products and the cash flow in that case was
as illustrated in figure 4, the differences between the initial cash and the cash available
by the end of the planning horizon was 3,757 EGP which is considered very low profit
43
cash flow for case 3
300000
250000
200000
150000
amount (EGP)
100000
50000
0
1 2 3 4 5 6 7 8 9 10 11 12
The model behavior was tested by changing influential parameters and checking the
response of the costs and revenues towards those changes. The chosen parameters were
uncertainties.
The parameters, in which their values were changed and tested, are the percent of
capacity occupied by MTO and MTS alternatively, fabric price changeability, the
5.1 Impact of changing percent occupied by MTO and MTS on the revenues. Since
for the base case, the percent of capacity occupied by MTO was 70 % and 30% for the
MTS production, then it was important to investigate the model sensitivity towards
44
varying percentages, when distributed alternatively between MTO and MTS. Table 23.
shows the different percentages tested and the corresponding gained revenues.
Table 19: Capacity distributed between MTO, MTS, and their projected revenues
Percent of capacity
occupied by
MTO/MTS production net profits MTO Rev. MTS Rev.
MTO MTS
70 30 153,227 1,433,100 559,000
60 40 294,263 1,240,000 793,000
50 50 478,107 1,030,000 1,200,000
40 60 509,472 827,000 1,240,000
Increasing the percent of capacity for MTS, the total revenues showed an increasing
pattern while the MTO revenue falls back as shown in figure 4. Although the revenues
attained in the graph makes sense, but that is not very accurate if the forecasted demands
were not fully sold as planned. On the other hand, the MTO shows a higher revenue
contribution when it occupied 40% of the capacity in contrary to the MTS when it
occupied 40% of the capacity. Moreover, analyzing the 70-30 percent alternatively
between MTO and MTS, when the former (MTO) occupied the higher percent,
significant revenues were depicted unlike the MTS, the model has reported infeasibility.
That was due to several things, the MTS production volumes are mainly restricted to the
cash availability, in which the MTO revenues contributes in it but on the other hand the
45
1600000
1400000
1200000
1000000
800000 Profits
The closer the percentage between the MTO and MTS capacity, while giving the MTO
Therefore, the most suitable percentages division would be 60-40 or 50-50 for allocating
5.2 Impact of distributing capacity between MTO and MTS on the costs.
The results for the costs encountered from MTO & MTS production during regular,
overtime and subcontracting show how the costs are sensitive to different production
46
40 60 350631 3.77E+05 82125 1555890
Figure 9 exhibits how the MTO production costs are high even if it occupies smaller
percent of the capacity. The subcontracting cost for the 50-50 and 40-60 percent were
apparently close with a difference of 7% and the total costs for the same percentages were
nearly the same. While when the MTS amounts exceeded that for the MTO; the 40-60
case; the total costs indeed decreased. However, when the MTS costs for the 40-60 and
50-50 case were compared, the difference in their costs will be just 0.2%.
2000000
1800000
1600000 fabric inv. costs
1400000 fabric costs
1200000
MTS inv. costs
1000000
MTO inv. costs
800000
subcontracting costs
600000
MTS prod. costs
400000
200000 MTO prod. costs
0
70-30 60-40 50-50 40-60
5.3 Impact of changing percent of capacity between MTO and MTS on the inventory
holding costs.
MTS, as the term implies, has higher inventory costs than that for MTO, table 25. As the
percent of capacity allocated for MTS increase the MTS inventory holding cost increases
notably as shown from the results in figure 5, except for the 70-30 %, the MTO had a
47
higher inventory, because it comprises higher percentage and there is enough storage area
Table 21: Percentage of capacity distributed between MTO and MTS vs. Inv. holding cost
MTO % MTS % MTO Inv. cost MTS Inv. cost
70 30 1785 216
60 40 1207 1433
50 50 1664 2237
40 60 648 4949
6000
5000
4000
3000
MTO inv. cost
2000 MTS inv costs
1000
0
70-30 60-40 50-50 40-60
Percentages of capacity consumed vs the inv. holding cost
Figure 10: Percentage of capacity distributed between MTO and MTS vs. inventory holding cost
Fabric price is a substantial parameter of the model. Thus measuring its implications on
the revenues and inventory cost was vital. Table 26, demonstrates the revenues and
inventory holding costs obtained for every percent increase in the fabric price.
Table 22: Fabric price percentage increase vs. profits and fabric inventory cost
Fabric price percent Profits Fabric inventory cost
increase
Base 153,226 26,921
48
5% 100,095 30,863
10% 61,509 36,353
15% 9,025 39,298
20% -45,082 41,453
25% -94,189 42,481
The profits were very sensitive to the fabric price increase as shown in figure 6; it showed
a decreasing pattern for every percent increase in the fabric price, which almost decayed,
and no revenues were realized when it increased by a percent more than 15 %. On the
contrary, to the fabric inventory cost that moved in an increasing fashion; which was
expected; as it is purchased and stored for the sake of the objective function optimality.
200000
150000
100000
50000
net profits
0 Fabric inventory cost
base 5% 10% 15% 20% 25%
-50000
-100000
-150000
Fabric price percent increase
Table 27, shows the impact of increasing the inventory holding cost for MTO, MTS and
fabrics on their costs and the revenues. The fabric inventory cost showed the higher
holding cost then the MTO and the least was the MTS inventory. That is consistent with
49
the data for fabric amounts for the MTO products. Since one garment of a MTO product
consumes 3 meters of fabrics unlike the MTS items, one garment of it never reach this
number. Given that the MTO production to MTS production is 70 to 30, then the amounts
Table 23: Profits and costs for inventory holding cost increase
Percent Profits MTO MTS MTO MTS Fabric
increase revenues inventory
revenues inventory inventory
costs
costs costs
Increasing the inventory holding cost by 10%, had a significant effect on the profits as it
decreased by 16%, while further increases in the inventory holding cost has decreased the
50
180000
160000
140000
120000
profits
100000
MTO inv. costs
80000 MTS inv. costs
60000 fabric inv. costs
40000
20000
0
base 10% 15% 20% 25%
cost and fabric inventory cost were both sensitive to the increase in the inventory holding
cost while the MTS weren’t significantly affected by that increase and that is because
The model considers subcontracting for the MTO products; hence, testing its impact on
the revenues and related MTO production costs was meaningful. Table 28 displays the
outputs for the revenues, MTO production costs, overtime costs, subcontracting costs and
MTO inventory costs that are merged for every 5 percent increase in the subcontracting
cost.
Table 24: Revenues and costs vs. subcontracting cost percentage increase
51
Percentage Profits MTO Overtime
Subcontracting MTO
increase Production production cost inventory
in the cost cost for cost
When the subcontracting cost increase the revenues decreased by around 4% for every
percent increase in the subcontracting cost. The MTO production cost was not
significantly affected by that increase; it was just elevated on an average of 0.2% and the
overtime production cost has remained almost unchanged. The inventory cost for MTO
Therefore, the revenues were sensitive to the changes in the subcontracting cost for the
percentages from 5 to 15 %, as presented in figure 13. A trial was made to test the
changes if the subcontracting cost increased by 20% but the solver took a long time to
solve (more than a day) to provide an optimum solution. So it was not included and for a
practical reason that this percent increase was considered an extreme and typically, the
subcontracting cost was not raised beyond the 10%, that is why the percentages tested
were enough.
52
Profits
155000
150000
145000
140000
135000
130000
125000
base 5% 10% 15%
An MILP model for production and inventory planning of mixed MTO – MTS system for
a readymade garment industry, is proposed. The developed model showed the successful
approach of having a MTS production along with the MTO production stream, as a new
way for overcoming financial drawbacks and the repercussions of depending solely on the
MTO demands.
The optimum results attained from the developed model helped at making the right
decisions regarding the inventory and production for different products for a mixed MTO-
MTS products. The model considered the vital decisions encountered from the impact of
The challenge faced at such environment was to guarantee the generation of a significant
revenue by the end of the season and positive cash flow at the end of each period. Thus, a
53
steady production process will take place without financial shortages. Cash inflow was
generated from the sales made by the MTS and the MTO items. The production plan
considered the main factors that influence the production process. The factors were the
fabric needed for production, the capacity limitations and the cash availability.
MTO due dates and forecasted demands were met for the planned products. Using the
spare capacity to produce to stock had a significant contribution in the revenues and
The model was very sensitive to the changes in the percentages of capacity allocation for
MTO and MTS to different cost parameters, while distributing the capacity 60 to 40
percent for MTO/ MTS products respectively proved to be the best option. For the given
MTO demands and target MTS, the tests on the percentage of MTS to be produced out of
the desired target production, has proved to present a positive cash towards the end of
Fabric price was a crucial parameter and the model was very sensitive to its changes,
which was anticipated, as the fabric price comprises 90% of the garment material cost.
Cash availability restricts the amounts to produce for MTS and affects the decision for
accepted orders. The results obtained helped at giving a practical guiding decisions that
The effectiveness of the model was distinct in its simplicity and applicability to real life
process. It has also tackled the vital and tangible decisions that happened to be of
54
The challenge has lied in having an optimum solution from merging the capacity and
production planning decisions with the finances in order to help a garment business to
sustain and grow. Therefore, a policy for production and capacity planning in garment
1. Consider accounting for lost sales for a better forecast and consider backlogging
3. Future work may also include examining the implications of the interest rate for
4. Adding to the model procurement decisions related to the fabric that may include
55
References
1. IBM Corporation. (2017). IBM ILOG CPLEX Optimization Studio. NY: IBM
3. Beemsterboer, B., Land, M., & Teunter, R. (2016), Hybrid MTO-MTS production
248(2), 453-461.
4. Carr, S., & Duenyas, I. (2000). Optimal admission control and sequencing in a
709-720.
5. Chen, J., Chen, C., Su, L., Wu, H., & Sun, C. (2012). Assembly line balancing in
6. Choy, K. L., Leung, Y. K., Chow, H. K. H., Poon, T. C., Kwong, C. K., Ho, G. T.
7. Ebadian, M., Rabbani, M., Jolai, F., Torabi, S. A., & Tavakkoli-Moghaddam, R.
(2008). A new decision-making structure for the order entry stage in make-toorder
8. Ebadian, M., Rabbani, M., Torabi, S. A., & Jolai, F. (2009). Hierarchical
56
short and reliable delivery dates. International Journal of Production Research,
47(20), 5761-5789.
for the Textile Industry with Make to Order and Make to Stock Considerations.
10. Gharehgozli, A. H., Rabbani, M., Zaerpour, N., & Razmi, J. (2008). A
12. Guo, Z. X., Wong, W. K., Leung, S. Y. S., Fan, J. T., & Chan, S. F. (2006).
Mathematical model and genetic optimization for the job shop scheduling
13. Hadj Youssef, K., Van Delft, C., & Dallery, Y. (2004). Efficient scheduling rules
14. Hemmati, S., Ebadian, M., & Nahvi, A. (2012). A new decision making structure
57
15. Kaminsky, P., & Kaya, O. (2009). Combined make-to-order/make-to-stock supply
16. Khakdaman, M., Wong, K. Y., Zohoori, B., Tiwari, M. K., & Merkert, R. (2015).
13581386.
17. Mok, P. Y., Cheung, T. Y., Wong, W. K., Leung, S. Y. S., & Fan, J. T. (2013).
18. Mok, P. Y. (2011). Intelligent apparel production planning for optimizing manual
operations using fuzzy set theory and evolutionary algorithms, the 103-110.
19. Morikawa, K., Takahashi, K., Hirotani, D. (2014). Make-to-stock policies for a
20. Ohta, H., Hirota, T., & Rahim, A. (2007). Optimal production-inventory policy
for make-to-order versus make-to-stock based on the M/Er/1 queuing model. The
21. Rafiei, H., & Rabbani, M. (2011). Order partitioning and order penetration point
58
23. Rajagopalan, S. (2002). Make to order or make to stock: Model and application.
251258.
25. Soman, C. A., van Donk, D. P., & Gaalman, G. (2004). Combined make-to-order
26. Soman, C. A., van Donk, D. P., & Gaalman, G. J. C. (2007). Capacitated planning
27. Ünal, C., Tunali, S., & Güner, M. (2009). Evaluation of alternative line
79(10), 908-916.
28. van Donk, D. P. (2001). Make to stock or make to order: The decoupling point in
69(3), 297-306.
29. Wang and Rosenshine (1983). Scheduling for a combination of made-to stock and
21(5):607.
30. Wong, W. K., Chan, C. K., & Ip, W. H. (2001). A hybrid flowshop scheduling
59
31. Yeh, C. (2000). A customer-focused planning approach to make-to-order
33. Zhang, T., Zheng, Q., Fang, Y., & Zhang, Y. (2015). Multi-level inventory
34. Zhang, Z., Kim, I., Springer, M., Cai, G., & Yu, Y. (2013). Dynamic pooling of
60
Appendix A: The inputs for multiples of MTO – MTS products
For multiple of products, Table A1 indicates the input capacities for 64 MTO and 128
MTS products. The distribution of the demands over the periods were distributed
arbitrarily.
Table A1: Input capacities for the 64, 128 MTO - MTS products
Capacities available per period
Table A2 represents the input costs used for the 64 MTO and 128 MTS products.
A1
Table A3: Input price for 64 - 128 MTO - MTS
MTO product Price (EGP) MTS product Price (EGP)
MTO product 1 170 MTS product 1 200
MTO product 2 110 MTS product 2 90
MTO product 3 90 MTS product 3 90
MTO product 4 90 MTS product 4 100
MTO product 5 80 MTS product 5 60
MTO product 6 80 MTS product 6 70
MTO product 7 170 MTS product 7 80
MTO product 8 110 MTS product 8 90
MTO product 9 90 MTS product 9 100
MTO product 10 90 MTS product 10 90
MTO product 11 80 MTS product 11 90
MTO product 12 80 MTS product 12 100
MTO product 13 170 MTS product 13 60
MTO product 14 110 MTS product 14 70
MTO product 15 90 MTS product 15 80
MTO product 16 90 MTS product 16 90
MTO product 17 80 MTS product 17 200
MTO product 18 80 MTS product 18 90
MTO product 19 170 MTS product 19 90
MTO product 20 110 MTS product 20 100
MTO product 21 90 MTS product 21 60
MTO product 22 90 MTS product 22 70
MTO product 23 80 MTS product 23 80
MTO product 24 80 MTS product 24 90
MTO product 25 170 MTS product 25 100
MTO product 26 110 MTS product 26 90
MTO product 27 90 MTS product 27 90
MTO product 28 90 MTS product 28 100
MTO product 29 80 MTS product 29 60
MTO product 30 80 MTS product 30 70
MTO product 31 170 MTS product 31 80
MTO product 32 110 MTS product 32 90
MTO product 33 90 MTS product 33 200
MTO product 34 90 MTS product 34 90
MTO product 35 80 MTS product 35 90
MTO product 36 80 MTS product 36 100
MTO product 37 170 MTS product 37 60
A1
C
MTS product 79 80
MTS product 80 90
MTS product 81 200
MTS product 82 90
MTS product 83 90
MTS product 84 100
MTS product 85 60
MTS product 86 70
MTS product 87 80
MTS product 88 90
MTS product 89 100
MTS product 90 90
MTS product 91 90
MTS product 92 100
MTS product 93 60
MTS product 94 70
MTS product 95 80
MTS product 96 90
MTS product 97 200
MTS product 98 90
MTS product 99 90
MTS product 100 100
MTS product 101 60
MTS product 102 70
MTS product 103 80
MTS product 104 90
MTS product 105 100
MTS product 106 90
MTS product 107 90
MTS product 108 100
MTS product 109 60
MTS product 110 70
MTS product 111 80
MTS product 112 90
MTS product 113 200
MTS product 114 90
MTS product 115 90
MTS product 116 100
MTS product 117 60
MTS product 118 70
D
MTS product 119 80
MTS product 120 90
MTS product 121 100
MTS product 122 90
MTS product 123 90
MTS product 124 100
MTS product 125 60
MTS product 126 70
MTS product 127 80
MTS product 128 90
E
2 436256 303177 57342 20725
3 436256 223560 53942 5992
4 330191 223574 48432 3303
5 306290 137660 43808 7242
6 253610 100218 29614 1808
7 245936 0 27814 0
8 149993 644 20612 40751
9 142169 50367 17712 26252
10 89813 32223 7712 13002
11 89600 863 0 3537
12 599 129151 0 0
5000,2 14 2 5000
2818,
5000,3
15 2,3 138 3 2044 2 2818
5000,4 16 4 521 4 4479
5000,5 17 5 5000
5000,6 18 5, 6 4917, 33 6 50 5 4917
F
5000,7 19 7 5000
5000,8 20 8 5000
2733,
6, 6,7-8,
3000,9 2733, 2993,
7, 9 9-12
21 260, 51 44
9,
3000,10 9 68
22 10 17, 2932 9 51
9,
3000,11 9, 10 2940
23 11 2940, 60
11,
3000,12 11, 12 71, 7
24 12 14, 2936 11 57
3000,1 25 1 3000
3000,2 26 1 2998 1 2 1 3000
3000,3 27 3 3000
3000,4 28 4 3000
3000,5 29 5 3000
3000,6 30 5 3000 5 3000
2-4, 56
2500,7 55, 109
31 2, 7 55, 2391 5 54
2500,8 32 7 2500 8 1 9 2500
2500,9 33 7 2499 9 1 10 2500
2500,10 34 9 2500
2500,11 35 10 2500
2500,12 36 12 2500
2500,1 37 1 2500
2500,2 38 1 2500 1 2500
2500,3 39 3 379 2 2121 2 2121
2500,4 40 4 2500
7000,5 41 5 670 5 6330
6,7….
1
7000,6 42 6 7001 12
7000,7 43 7 6999 5 1 5,6 1
7000,8 44 8 7000
7000,9 45 9 7000
7000,10 46 10 7000
6-7, 8-
6999,70
7000,11 47 6, 8 6999, 50 10,11-
49,49
12
7000,12 48 12 7000
7000,1 49 0 1 7001 1 …12 1
7000,2 50 2 7000
9000,3 51 3 9000
9000,4 52 4 9000
G
9000,5 53 5 8584 5 416
9000,6 54 5, 6 1, 2417 5, 6 49, 6533 5 50
9000,7 55 7 9000
9000,8 56 8 8942 7 58 7 58
9000,9 57 9 9000
8,
6430,
9, 6430, 8, 9
6480
9000,10 58 10 50, 2520
10, 2966,
10 2966
9000,11 59 11 6034
9000,12 60 12 9000
7000,1 61 1 7000
112,
1 113
7000,2 62 1, 2 6887 1 1
7000,3 63 3 6999 3 1
7000,4 64 4 7000
600, 1 16 0 0 0
2000,3 17 0 10 2000 10, 11 2000, 2000
2000,3 18 12 2000 0 0
H
2000,3 19 10 2000 0 10,11 2000, 2000
2000,3 20 0 11 2197 11, 12 2197, 197
2000,3 21 0 8 2000 8,9..12 2000
2000,3 22 11 2000 0 11 2000
2000,3 23 0 0 0
2000,3 24 11 2000 0 11 2000
900,3 25 2 900 0 2 900
900,3 26 0 3 900 0
900,3 27 0 3 900 0
900,3 28 0 3 900 0
900,3 29 2 900 0 2 900
900,3 30 0 3 900 0
900,3 31 2 900 0 2 900
900,3 32 3 900 0 0
700,2 33 0 1 700 1 700
700,2 34 0 1 700 1 700
700,2 35 2 700 0 0
700,2 36 0 700 1 700
700,2 37 2 700 0 0
700,2 38 2 700 0 0
700,2 39 1 700 0 1 700
700,2 40 2 700 0 0
1499, 6-7, 8-9, 1499, 1549,
1500, 10 41 6, 8 0
50 10-12 49
1500, 10 42 0 8 1500 8,9 1500, 1500
1500, 10 43 10 1500 10 1500 0
1500, 10 44 0 0 8,9 1500, 1500
1500, 10 45 0 0 0
1500, 10 46 8 1500 0 8,9 1500, 1500
1500, 10 47 9 1500 0 9 1500
1500, 10 48 9 1500 0 9 1500
700,3 49 2 700 0 2 700
700,4 50 2 700 0 2 700
700,5 51 2 700 0 2 700
700,6 52 2 700 0 2 700
700,7 53 3 700 0 0
700,8 54 2 700 0 2 700
700,9 55 1 700 0 1,2 700,700
700,10 56 1 700 0 1,2 700,700
1100, 6 57 6 1100 0 0
I
1100, 6 58 2 1100 0 2,3..5 1100
1100, 6 59 2 1100 0 2,3..5 1100
1100, 6 60 1 674 2 426 1, 2-5 674, 1100
1100, 6 61 6 1100 0 0
1100, 6 62 1 1100 0 1,2..5 1100
1100, 6 63 0 2, 5 212, 888 2-4,5 212, 1100
1100, 6 64 5 66 6 1034 5 66
1500, 10 65 10 1500 0 0
1500, 10 66 10 1500 0 0
1500, 10 67 10 1500 0 0
1500, 10 68 0 10 1500 0
1498, 6-7,8- 1498, 1549,
1500, 10 69 6, 8 0
51 9,10-12 49
1500, 10 70 10 1500 0 0
1500, 10 71 9 1500 0 9 1500
1500, 10 72 10 1500 0 0
1000, 3 73 1 1000 0 1,2 1000, 1000
1000, 3 74 3 1000 0 0
1000, 3 75 0 3 1000 0
1000, 3 76 3 1000 0 0
1000, 3 77 2 1000 0 2 1000
1000, 3 78 3 1000 0 0
1000, 3 79 2 1000 0 2 1000
1000, 3 80 1 1000 0 1,2 1000, 1000
1000, 4 81 3 1000 0 3 1000
1000, 4 82 3 1000 0 3 1000
1000, 4 83 2 1000 0 2, 3 1000, 1000
1000, 1000,
1000, 4 84 1 1000 0 1,2,3
1000
1000, 4 85 4 1000 0 0
1000, 1000,
1000, 4 86 0 1 1000 1, 2, 3 1000
1000, 4 87 0 3 1000 3 1000
1000, 4 88 0 2 1000 2, 3 1000, 1000
1300, 5 89 5 1300 0 0
1300, 1300,
1300, 5 90 2 1300 2, 3, 4 1300
1300, 1300,
1300, 5 91 2 1300 0 2, 3, 4 1300
1300, 1300,
1300, 5 92 2 1300 0 2, 3, 4 1300
288, 288, 288, 288
1300, 5 93 2, 5 1010 5 2 2, 3, 5
J
1300, 5 94 3 1300 0 3,4 1300, 1300
1300, 5 95 0 3 1300 3,4 1300, 1300
1300, 5 96 0 5 1300 0
1300, 6 97 0 4 1300 4, 5 1300, 1300
1300, 6 98 0 4 1300 4, 5 1300, 1300
1300, 6 99 5 1000 0
1300, 6 100 5 1000 0
1300, 6 101 5 1000 0
1300, 6 102 5 1000 0
1300, 6 103 5 1000 0
1300, 6 104 0 2 1000 2,3…5 1000
900, 7 105 0 6 900 6 900
900, 7 106 0 7 900 0
900, 7 107 0 7 900 0
783,
900, 7 108 3, 7 0 3,4…6 783
117
900, 7 109 7 900 0 0
900, 7 110 0 7 900 0
900, 7 111 0 7 900 0
900, 7 112 6 900 0 6 900
2000, 12 113 11 2019 0 11, 12 2019, 19
2000, 11 114 10 2000 0 10 2000
2000, 10 115 10 2000 0 0
2000, 11 116 9 2000 0 9, 10 2000, 2000
2000, 12 117 12 163 11 1837 11 1837
2000, 11 118 9 2000 0 9, 10 2000, 2000
2000, 12 119 12 308 11 1692 11 1692
2000, 11 120 0 11 2000 0
1500,9 121 0 8 1500 8 1500
1500,9 122 0 9 1500 0
1500,9 123 0 9 1500 0
1500,9 124 0 9 1500 0
1500,9 125 9 1500 0 0
1500,9 126 9 1500 0 0
1500,9 127 9 1500 0 0
1500,9 128 0 8 1500 8 1500
K
Appendix C: Fabric inventory and quantity purchased for the 32 – 64
MTO – MTS
Table C1: fabric inventory for 32-64 MTO - MTS
IFft
1 2 3 4 5 6 7 8 9 10 11 12
287599 244456 233956 160471 145381 103849 100762 43306 22960 24250 250 250
1151 39494 24544 21029 1369 17849 1686 0 0 0 0 0
17600 11600 4400 4400 0 2 2 5898 0 0 0 0
43150 37200 24350 23350 22250 19050 19050 18796 18540 2000 0
1 2 3 4 5 6 7 8 9 10 11 12
0 0 0 0 0 0 0 0 0 30 0 0
0 0 0 0 0 0 0 0 0
0 0 0 0 0 2 0 100 100 0 0 0
0 0 0 0 0 100 0 0 0 0 0 100
1 2 3 4 5 6 7 8 9 10 11 12
296956 0 0 0 0 0 0 0 0 1260 0 0
65636 57720 18152 5485 35945 52033 65575 0 0 0 0 0
20000 0 0 0 0 0 0 6122 2 0 0 8000
46450 0 0 0 0 0 0 0 0 0 0 100
A1