Supply Chain System of Systems

AAE560

System of Systems Modeling and analysis

Final Project

Supply chain for Large Monolithic systems (Team -2)

Submitted by,

Akul Goyal

Keshav Amburay

Naga Tejadeep Kalipi

Yuvraaj Kannabiran Sathyanarayanan

INTRODUCTION Air travel is the primary mode of transportation for travelling large distances. Current generation passenger aircrafts as manufactured by Boeing, Airbus and others, are highly complex machines that have various sub-systems that work in synergy to safely transport people and goods from one place to another. Original Equipment Manufacturers (OEM) do not manufacture all of these sub-systems but rather contract the manufacturing operations to suppliers. Each of these entities (OEMs and Suppliers) have their own management, their own independent systems of operations and have their own policies and procedures. Hence, each entity is a system in its own right and the overall manufacturing and supply chain system is therefore a System of Systems. PROJECT BACKGROUND The effect of introduction of new technology on aircraft manufacturing supply chain is analyzed using System of Systems approach. This study is restricted to study of impact of utilization of composite material in major aero structures on the aircraft supply chain system. Use of composite material has been rising steadily in aircrafts. Figure 1 shows the percentage of usage of composite materials in different sectors of aerospace industry. Usage of composite material in aircraft design will potentially give rise to the need of new suppliers with advanced manufacturing capabilities. The reason for this increase is that 30 percent of composite structures are manufactured with automated production equipment and it is predicted it would be 80 percent in the next 10 years from the cost saving and increased throughput rate attributed to robotic systems.

PROJECT OVERVIEW The overview of this project is to understand and analyze the aircraft manufacturing supply chain. It also helps to quantify the emergent behavior due to delaying effect in manufacturing a mainstream product by incorporating a new technology. The change and flexibility in the supply chain by re-examination of the capabilities of the suppliers, distributors and warehouses and to check their capabilities to deal with the new technology with same or better efficiencies as before and possible restructuring of the supply chain itself. Also adding to this there is an uncertainty in quantifying the efficiency in replacing existing product. The figure 2 from the national institute for aviation research shows the supply chain transformation from 1980’s to current scenario highlighting the changes in the interactions between OEMs, suppliers and sub-suppliers from 1980s to the present day.

Figure 2. Commercial Aircraft OEM – Composite

supplier relationships

Figure 1.Usage of Composite Materials in different sectors of aerospace industry

BOEING 787 PROGRAM The Boeing 787 aircraft structure is 80% & 50% composite by volume and weight respectively. The selection of these composite material and their usage criteria heavily depends on fuel efficiency by reducing aircraft weight & using fuel-efficient engines. In addition, it should be noted that Boeing outsource 70% of the parts in order to build its aircraft. Figure 3 highlights the major sections of the 787 aircraft, its manufacturers and their location.

Figure 3. Boeing 787 aircraft structure assembly breakdown

Some of the issues faced by Boeing 787 program are listed below:

First delivery was nearly 3 years late from promised • The cost of producing a 787 exceeded the purchase price at the end of 2013 • Boeing lost $30 million per 787 delivered in the first quarter of 2015 • The 787 program is expected to be profitable after 1,100 aircraft have been sold • Current Order State: Total received – 1105, Total delivered – 258

The aim of this project is to model aircraft manufacturing supply chain as a system of systems. The project is divided into three phases viz. Definition, Abstraction and Implementation. DEFINITION PHASE This report deals with the formulation of a SoS problem a la the Definition Phase. This step is key towards problem development and modelling as it builds a strong base by means of a well-defined problem. It ensures that the problem fits the characteristics of the SoS by defining its operational context, status quo, barriers and the SoS lexicon. In effect, these help examine the traits of the SoS problem, which is vital. OPERATIONAL CONTEXT Commercial Supply Chain has since long been considered a challenge to manage, its implications on large monolithic production system such as airplanes proves all the more difficult. Such supply chains attend to Air Transport domains which in turn serve various stakeholders such as nations, governments (military), private organizations as well as public sectors (general day to day commuters). These systems incorporate very complex networks, which consist of different levels of suppliers, production plants, warehouses, distribution branches and retailers (OEM), through which the raw materials are transformed into the final product to be delivered to the customer. It should be noted that since these supply chains are formed to serve the needs and the demands of the consumer, in certain cases the products are produced on a made-to-order basis with stringent delivery deadlines and cost factors involved.

STATUS QUO The supply chain systems in operation till today at their grassroots level (β and below levels) are built on the assumption of a stable and high volume production throughout time, taking leverage of the low cost labor in developing countries (Malik et al., 2011). The uncertainty in these systems may be caused by a multitude of factors some of which, being faced by the small-part large-volume supply chain systems are:

Political disturbances in one country can affect the whole production due to global nature of

supply chain and can have major implications if the primary supplier is present in that region.

Economic volatility present in developing countries (where most of the production of the small

scale components happens) and the rise in income levels causing the emergence of new markets

there have also added to the complexity of the supply chains as a whole and have caused

uncertainty in their behavior (Malik et al., 2011).

The increased probability and frequency of technological disruptions and rapid development of their markets and suppliers requires a very dynamic supply chain formation and management processes at the OEM level.

However, in case of large monolithic systems, these systems face some unique issues such as:

Effect of Flexibility of Consumer Demand: Manufacturers purely design and develop on the primary structure of the airplane, which cannot be altered, and collaborate with the airline company on specific traits they may have to request for which they have to deal with required suppliers. Sometimes while processing these requested traits, the airline may revert back and ask for some changes to their demand due to change in certain drivers such as operations (flight routes) or regulations (government restrictions). This change in demand is very likely to affect the supply chain that the manufacturer has established.

Introduction of new technologies: Technological advancements may change the design of the products and the way in which they are realized, in the industry. This may lead to a restructuring of the supply chain due to the addition/removal of suppliers and the change of hierarchy in between them. It involves re-examination of the suppliers, distributors and warehouses for their adaptability with the technology and possible restructuring of the supply chain itself.

BARRIERS Uncertainty due to the high complexity present in these supply chains and their interdependence on

several highly dynamic aspects like the economy is the primary cause of concern as this leads to

uncertainty in their operation. The key to understanding it would be to know the interactions between

various systems present in it and to model the disruptors and stakeholders in a SoS framework. This would

lead to an understanding of the mechanisms causing the behavior and the conditions of the future states

of the system.

CURRENT TRAITS OF THE SOS The table 1 shows the current traits of the SoS problems in lieu with the traits of SoS observed and offered by Maier and DeLaurentis.

Table 1. Traits of Aircraft Manufacturing/Supply Chain System of Systems

Traits Description

Operational &

Managerial

Independence

Lower level elements in the supply chain have their own purposes outside the SoS and operate

independently as well. At the higher levels of the supply chain, they do tend to exhibit own purposes

and but show a sense of dependence to operate to fulfill them

Geographic Distribution It is evident that the constituent are geographically distributed hence the concept of Supply chain exists

which relies on their necessary coordination to work towards the common goal of the SoS

Evolutionary Behavior The challenges that supply chains pose come from not only their complicated networks but also their

dynamic nature of evolutionary behavior wherein various elements are added or get removed during

the process

Emergent Behavior With different individual purposes that the constituents have, there is an certain level of expected

emergent behavior with some level of uncertainty in the behavior also thought to exist

Heterogeneity The stakeholders involved may not be of a different nature in entirety but are of different dynamics and

time scales based on their operational limitations and capabilities

Networks of Networks Interactions between these individual systems result in connectivity between them. This connectivity

highlights the nature of the interactions

Trans-domain Unquestionably a trans-domain environment requiring the amalgamation of knowledge from various

backgrounds such as engineering, policy making, operations, planning, economics and more

PROBLEM SCOPE & ROPE TABLE The scope of the problem within supply chain domain can be well described using a ROPE table, which is highlighted in the table below.

Table 2. ROPE Table & Problem Scope

Level Entities Resources Operations Policies Economics

α Individual Entities People, Machinery,

Products, Parts

Machine level manufacturing, Maintenance of machines and

systems

Standard Operating Procedures, HR Policies, Safety

policies

Wages, Maintenance Costs, Component costs

β Teams/Departments Shop floor, Quality

Control, HR Department

Production planning, R&D, Quality Testing

Engineering Specifications, R&D

policies, Quality Assurance policies

Departmental Budget, Expenditure

γ Manufacturing and

Testing Facilities

Collection of all the departments, IT

systems

Manufacturing, Production Level

tracking and management,

Prototyping, Field Trials, Data Sharing

Energy policies, Waste handling policies, Data management

policies

Revenue and Expenditure, Power

usage charges, utility charges

δ Single OEM or Supplier Aircrafts, Finance Team, Company

Board

Sales and Customer Service, Regulatory

Compliance

Profit/Loss sharing agreements

Costs of Manufacturing, Testing, Profit/Loss,

Cash flow

ε Market

Airlines, OEMs, Suppliers, FAA,

IATA, Regulatory agencies

Buying and Maintenance of

Aircraft, Enforcement of Regulations, Joint

design of Aircraft

Customer/OEM agreements,

Regulatory policies, Safety policies

Profitability, Cash flow, Budget

ABSTRACTION PHASE

The second phase of the project involved identification of the main stakeholders of the SoS, the driving

factors and entities, disruptors and resources and interdependency networks between them.

The following entities form the primary stakeholders of the SoS:

1. Original Equipment Manufacturers (OEM)

The Original Equipment Manufacturer is the final manufacturer of the aircraft and all the technical

parameters and overall manufacturing strategy based on inputs from the airline (customer) and supplier.

2. Suppliers

The aircraft has various components not manufactured by the OEM itself but rather contracted to a third

party manufacturing entities. In the current SoS model, the term supplier is used to denote only the

suppliers who are directly conducting business with the OEM involved in manufacture of aircraft wings.

The key behavior and interests of the stakeholders of the SoS are shown in the paper model in figure4.

Figure 4. SoS Paper Model showing key behaviors of stakeholders

As shown in the paper model in Figure 4, the primary functions of the OEM is manufacturing, assembly

and delivery of aircraft to the airline. The primary functions of the supplier is manufacturing of the

required parts within the specifications defined by the OEM and timely delivery of manufactured parts to

the OEM. The supplier group denotes the suppliers who are currently doing business with the OEM. The

suppliers outside of this group are potential candidates to be an approved supplier for the OEM.

The operations of the airline listed in the paper model are specific to its primary function as an air-

transport service provider to the end-users (passengers). The regulatory agencies define and enforce the

broader level guidelines with respect to environmental concerns, passenger safety, technical and design

aspects of the aircraft.

The primary network connections between the entities in the supply-chain SoS are: 1. Airline OEM 2. OEM Supplier (within the approved suppliers group)

The interactions and connectivity between the stakeholders, resources, drivers and disruptors is shown in the model in figure 5.

Figure 5. Interactions between stakeholders, resources, drivers and disruptors

Figure 6. System Dynamics model of interactions in the Supply Chain SoS

From the system dynamics model in figure 6, it can be seen that the introduction of the technology involves transfer of lots of between links connecting to the classes of the system (OEM or Supplier). Within these classes, parametric information such as design requirements and performance requirements are transferred and delegated accordingly. Policies are generally defined and enforced by Regulatory/Government agencies and all types of entities (Airlines, OEMs and Suppliers) have to abide by those policies.

CONTROL The aircraft manufacturing supply chain system-of-system consists of entities, which are not controlled by a supra-entity. Each of the constituent entity of the SoS has its own objectives, own management, authority, organization structure (may or may not be distinct from other entities), internal objectives and policies, financial management systems etc. The activities at the overall SoS are not managed by a SoS level central authority but rather each entity manages its activities in accordance with its contractual obligation to other entities. Hence, the aircraft manufacturing supply-chain system is a Collaborative SoS from an autonomy perspective. IMPLEMENTATION PHASE

Our objective was to build a model of the SoS to simulate the time evolution of the supply chain’s

capabilities in the implementation of new technology in aircraft manufacturing as a mainstream product.

This is done by not only modelling and analyzing the flexibility of the supply chain, but also the effect on

the lead times in terms of new product developments and improvement of efficiency. Time and

production output are the parameters that define a Supply Chain system. However, it is due to many other

factors such a capabilities, efficiencies, cost, etc. that these simple parameters cannot be easily predicted

having known the configuration of a particular supply chain.

HYPOTHESIS

We proposed two hypothesis for the SoS. One was a simple yet hard to predict notion that reduction in lead-time could be directly related to increase in investments towards developments. The second however was based on an opinionated theory. It was that the flexibility of substitution of suppliers other than those contracted for manufacturing the product, for a small nominal cost, can result in substantial reduction of time delays with a rise in the supply chains outputs towards the desired results. DESIGN VARIABLES (PERFORMANCE METRICS)

Three design variables were used to study the model: (1) assimilation time, (2) upgradation cost, and (3)

throughput rate of the supply chain. The assimilation time in this scenario is the time required to obtain

the desired outputs for reaching a break-even point on profitability on investments. It could also be seen

in the form of delays encountered to attaining the goals. Upgradation cost is seen as the investment that

were to be made towards the required reduction in delays. Throughput rate which is a common

terminology used in Production and supply chain merely refers to the number of aircrafts delivered (per

month) to satisfy the loss gap faced by the company under the program.

MODELLING APPROACH

There were numerous simulation and modelling approaches available, which could be used for this

application. These were namely Agent based modelling, System dynamics and Discrete event simulation.

Since discrete event simulation though applicable to supply chain environments, is not capable of working

with the constituents of the SoS as agents which interact with each other, it was ruled out. It relied more

on the concept of sequencing and iterating a set of well-defined events. Initially, Agent based modelling

had been selected for the purpose of modelling the SoS, but due to issues faced (discussed later) with this

approach, the team had to resort to using the system dynamics approach using causal feedback loop

technique.

MODELLING

The modeling of the SoS was done in Simulink. Initially, an agent-based model was selected to be created

and AnyLogic was used for that. But later, it was discovered that the modeling of current configuration of

SoS was quite difficult in it and hence, the focus was shifted to Simulink, which is a symbolic computing

interface for the MATLAB.

The model of SoS has been constructed on three levels. The top most level is that of the OEM and the

level below it is of the suppliers. The level below the suppliers is that of the sub-suppliers. In the current

configuration, the OEM was assumed to be receiving from 3 suppliers and the 3 suppliers themselves had

3 sub-suppliers for each of them. The sub-suppliers of one supplier were not related to or affected by the

sub-suppliers of another supplier.

The efficiency and the productivity of each supplier was dependent on his sub-suppliers in a direct

manner. It was assumed that the sub-suppliers` efficiencies were directly influenced by the market and

the external factors. Hence, the efficiency of sub-suppliers was found to be varying on a random Gaussian

distribution with the parameters mean and standard deviation, which were adjusted for each sub-supplier

such that the total efficiency at any point of time was within the bounds demonstrated by independent

real world manufacturing units.

The model has been built such that on every time step, the OEM selects the supplier with maximum

efficiency. Since the suppliers’ efficiency is determined by its sub-suppliers and its productivity, and the

efficiency of the sub suppliers varies according to a Gaussian distribution, the total efficiency of each

supplier also varies over a wide range throughout the time.

The two parameters used to measure the system performance are the efficiency of the group of suppliers

(the max efficiency of the suppliers) and the no. of parts produced by the supplier. According to the

hypothesis, the SoS behavior has been studied for only one supplier and one primary supplier with two

others on standby. The model was run for three different investment scenarios: 6 Billion investment and

7 Billion investment, another set of 10 Billion investment and 11 Billion investment.

The following images show some of the Simulink models of the supplier behavior model, sub- supplier

network and the OEM level supplier network.

Figure 7. Supplier behavior model

Figure 8. Supplier dependency on sub-suppliers

Figure 9. Overall SoS at the OEM level.

RESULTS

The simulation was run for a time period of 20 seconds with a time step of 0.1 second. Each second in the

simulation is one year in the real world. It was observed from the simulation that having two suppliers on

standby increased the overall system efficiency and had a constant manufacture of parts required for the

airplane. When the primary supplier had experienced failure owing to the failure of his sub-suppliers,

other suppliers took over and there was a steady supply of the parts required to manufacture the airplane

and the overall system efficiency was maintained above 80% in almost all cases except 1. This was the

case where all the suppliers had failed together. In the real world, this is a very unlikely scenario as we

can assume that due to the global nature of supply chains for large monolithic systems, there is a

significant geographical spread of its supplier base. Hence, it should be safe to assume that the probability

of failure of all the suppliers at same time is quite low. In the simulation also, this was observed only at

one time step out of many. The following images show the results obtained through the simulation for

the investment of 6 Billion for only 1 supplier and 7 Billion for two standby suppliers.

Figure .10 Total efficiency of group of suppliers – 6 Billion investment Figure 11.Individual efficiency of suppliers with 6 Billion investment

Figure 12. No. of aircraft delivered- Best of the group – 6 Billion investment

Figure 13. Aircraft delivered-Individual supplier – 6 Billion investment

Similar results were obtained for other values of investment into the system. We observe that for one

particular investment value, the overall efficiency when a group of suppliers is present is always better

than having an individual supplier. The implications of these results will be discussed in the next section.

DISCUSSION

Now we would be discussing the results to see if they support our proposed hypothesis. The graph in

Figure 14. Supports our first hypothesis. Even if we maintain a single supplier in the system, an increment

in the investments made from the baseline value to the highest permissible value, we see a big jump in

the production value ultimately resulting the decrease of delays consequentially reducing the loss gap.

Figure 14. Effect of increase in investments to the output of the supply chain

The blue line represents the baseline investment of $6b whereas the red line is for a $10b investment. As

expected, there is a significant increase in the production rate of the supply chain, which would allow the

supply chain to recover from the loss gap much more quickly. There are these uncertain dips noticed with

a $10b investment where the output is actually lesser than what the supply chain produces with a $6b

investment. The authors were unable to determine the reason for this behavior. However, this could be

related to how the model assigns random variations of efficiency.

Figure 15, on the other hand supports the second hypothesis based on the improvement in the production

rate by having standby suppliers, who would substitute for the main supplier if in case it failed to meet

requirements. The setup of these standby suppliers involved a nominal agreement cost of $500m per

supplier. Since the model was developed with three suppliers in total, the range consisted of two standby

suppliers with an addition of $1b to the investments.

0

2

4

6

8

10

12

14

16

18

20

0 2 4 6 8 10 12

# o

f A

ircr

afts

del

iver

ed/m

on

th

Time period (years)

Supply Chain Output

$6b investment $10b investment

Figure 15. Supply Chain output impact w/ additional suppliers

It can be seen that by just having 2 suppliers on a standby state with a nominal investment fee of $1b, the

production rate of the supply chain becomes more stable and reliable maintaining a delivery of 12-13

aircrafts per month. A similar pattern is seen when comparing effects of $10b investment to $11b.

However, a thing to note here is that having more suppliers on a standby state does not actually increase

the production rate of the supply chain but improve the efficiency of the supply chain and stabilize its

production rate to the required. Possibly, having more suppliers may result in the increase of the

production rate, but that would require impact the required investment significantly.

ASSUMPTIONS

To reduce the complexities that would be faced during the modelling and simulation phase, the following

assumptions were to be made:

• The individual penalties to customers were not included in the total loss (calculation of loss gap)

incurred due to delays as they were considered negligible

• Since current investment data from Boeing would not be readily available due to legal issues, we

had to take past data and consolidate it into the present and future calculations for estimations

on the investments (upgradation costs)

• Likewise, to have each extra supplier on a ‘standby’ for substitution in case the originally selected

supplier failed, an assumed agreement value (nominal cost) of $500m was assigned per supplier

• Outsourcing of a majority of the manufacturing processes lead to difficulties in obtaining the

relevant information on their detailed processes leading to various assumptions as well which

virtually removed the outsourcing notion from the model and treated the whole situation as a

direct set of relationships between the OEM, suppliers and the sublevel suppliers.

0

2

4

6

8

10

12

14

0 2 4 6 8 10 12

# o

f A

ircr

afts

del

iver

ed/m

on

th

Time period (years)

Supply Chain Output

$6b investment (1 supplier) $7b investment (2 standby suppliers)

EMERGENT PROPERTIES

In reality, System Dynamics approach does not help capture emergent behavior but it can be credited to

the model if it can capture a glimpse of it. For this model, something similar seems to happen. Studying

the results it can be seen that the system evolves and tries making itself more robust in the manner that

it recovers from the failure of one supplier by making a switch to another more efficient supplier at the

next time step itself. This shows an evolutionary behavior of the model to adapt and survive.

Nevertheless, we would like to consider working on an agent based modelling incorporating more

intricate agent models with dynamic relations, which would allow us to study the emergent behavior in

the system better.

CONCLUSIONS AND FUTURE WORK PROSPECTS

In conclusion, it has been found that SoS approach is possibly a best solution to studying complex systems

such as supply chain networks involving interactions between different entities with different traits.

Nonetheless, it does not guarantee the true solution as it only provides scenarios and aftermaths which

help the decision makers take actions by assessing the compromises. Models made using SoS approach

may not be able to mimic real life situations.

In terms of future work, we intend to enhance and develop the model predominantly to have the possibility of the OEM connect to more than one supplier agents, which in anticipation could improve the dynamic behavior of the supply chain. We would like to introduce the complexity metric which is also very helpful in the decision making process. We were not able to append disruptor effects associated with one of the drivers, regulatory authorities. We could include parameters such as Safety rating and Environmental rating which are likely to effect the production rate of the supply chain. Lastly, we would like to introduce a system for failures of any of the nodes, which would allow for the system to become more robust in similarity to a Scale-free network. REFERENCES

[1] Axelrod, R. M., “The complexity of cooperation: Agent-based models of competition and

collaboration”, Princeton University Press, 1997

[2] Böhme, T., Deakins, E., Pepper, M. & Towill, D., “Systems engineering effective

supply chain innovations”, International Journal of Production Research, 52:21, 6518-6537, 2014

[3] Cizmeci, D., “An Examination of Boeing’s Supply Chain Management Practices within the

Context of Global Aerospace Industry”, Master’s Thesis, Massachusetts Institute of Technology, 2005

[4] Cohan, P., “Boeing's Dreamliner Delays: Outsourcing Goes Too Far”, Jan 2011,

http://www.dailyfinance.com/2011/01/21/boeing-dreamliner-delays-outsourcing-goes-too-far/

[5] DeLaurentis, D., “Understanding transportation as a system-of-systems design

problem”, In 43rd AIAA Aerospace Sciences Meeting and Exhibit (Vol. 1). Reno, NV. New York: AIAA, Jan

2005

[6] Holmberg, G., Strindberg, G., & Saab, A. B., “Manufacturing as a Multilevel System, Suppliers

and Supply Chain. & Proceedings”, 2003

[7] Kang, T., Lewe, J. H., & Mavris, D. N., “Uncertainty Dynamics and Economic

Modeling of the Air Transportation Supply Chain”, In 6th AIAA Aviation Technology, Integration and

Operations Conference (ATIO), Wichita, Kansas, Sep 2006

[8] Kurtenbach, E., “Boeing 787 supplier Mitsubishi sets sights on niche in aircraft sector”, Apr 2015,

http://www.postandcourier.com/article/20150410/PC05/150419877/1177/boeing-787-supplier-

mitsubishi-sets-sights-on-niche-in-aircraft-sector

[9] Lucintel, “Value Chain Analysis of Boeing 787: 2011-2016”, Research & Markets, Oct 2011

[10] Maier, M. W., “Architecting Principles for Systems‐of‐Systems”, In INCOSE International

Symposium (Vol. 6, No. 1, pp. 565-573), Jul 1996

[11] Malik, Y., Niemeyer, A. and Ruwadi, B., “Building the supply chain of the future”, Web article by

McKinsey and Company, 2011,

http://www.mckinsey.com/insights/operations/building_the_supply_chain_of_the_future

[12] MHI to increase Production of Composite Wing Boxes for Boeing 787 Dreamliner, Aug 2014,

http://www.mhi-global.com/news/story/1408071825.html

[13] Ostrower, J., “Boeing's Key Mission: Cut Dreamliner Cost”, Wall Street Journal, Jan 2014,

http://www.wsj.com/articles/SB10001424052702304361604579292333739357034

[14] Pinon, O. J., Fry, K., & Clarke, J. P., “The Air Transportation System as a Supply Chain. &

Proceedings”, 2009

[15] Suga, M., “Boeing, Toray in Talks to Expand $6 Billion 787 Deal”, Nov 2007,

http://www.bloomberg.com/apps/news?pid=newsarchive&sid=aRu6WIrqhc0Y

[16] Tang, Z. E., Pinon, O. and Marvis, D., “Identification of Key Factors in Integrating Aircraft and the

Associated Supply Chains during Early Design Phases”, School of Aerospace Engineering, Georgia

Institute of Technology, Atlanta, GA, June 16-20, 2014

[17] Tang, Z., Pinon-Fischer, O. J., & Mavris, D. N., “Identification of Key Factors in Integrating Aircraft

and the Associated Supply Chains during Early Design Phases”

[18] Tiwari, M. K., Mahanty, B., Sarmah, S. P., “Modeling of Responsive Supply Chain”, Page 54, August

20, 2012

[19] Tomblin, J., “Overview of Composite Material Trends in Aviation Manufacturing”, National Institute

for Aviation Research, Wichita State University

[20] Trimble, S., “Boeing 787 unit loss declines, but deferred costs rise”, Apr 2015,

http://www.flightglobal.com/news/articles/boeing-787-unit-loss-declines-but-deferred-costs-rise-

411502/

[21] Vinarcik, M. J., “Chapter 23: Airbus A380 and Boeing 787 Contrast of Competing Architectures for Air Transportation”, Case Studies in System of Systems, Enterprise Systems, and Complex Systems, Jul 2014