Koren Y. The Global Manufacturing Revolution: Product-Process-Business Integration and Reconfigurable Systems
Подождите немного. Документ загружается.
from France). The range of respondents covered a wide spectrum from Vice
Presidents and Executive Directors in the auto industry, to R&D Managers and
system designers in the machine tool industry.
The objectives of the survey were to
.
Understand the reasons for success and failures of FMS in the mechan-
ical industry.
.
Evaluate possible technologies that may contribute to the success of FMS.
The system configuration types utilized by the surveyed industry, as reported to us,
are shown in Figure 6.15.
Regarding the system design phase, the survey conclusions were
.
New flexible systems should be designed to produce both existing and future
products.
.
In many cases, the purchased systems had over-capacity and excess function-
ality. Purchasing over-capacity is more common than purchasing over-
functionality. That extra capacity was never utilized in 20% of all cases, and
the extra functionality was never utilized in 30% of them.
The time to design and install (ramp up) large manufacturing systems is considered
very critical by most experts. Regarding the system utilization phase, the survey
concluded that:
.
In most cases systems were utilized for the activities they were purchased for,
indicating coherence between design and use.
.
High satisfaction with the FMS was repor ted in all cases in which systems were
utilized to increase the product variety.
Figure 6.15 Types of system configurations reported in the survey.
INDUSTRY EXPERIENCE WITH FMS—A SURVEY 171
.
The reports were equally divided between cases in which syst ems are operated as
expected and cases in which systems were not to operate at their full potential.
The main conclusions of the survey were
System Cost is by far the most important factor in future success of large
flexible manufacturing systems. The main cost factors are (ranked in the order of
importance)
.
Maintenance cost
.
Floor space
.
Capital investment cost
.
Manpower cost to operate the systems
System Reconfiguration capability and ease of upgrading system capacity is
the second most important issu e.
Workforce Training is extremely important to effectively operate FMSs.
Machine and control reconfiguration capabilities are important features, but
they are at lower priority than the items listed above .
The major findings of the survey are summarized below.
.
Respondents were equally divided whether FMSs are being operated at their
potential performance (41% YES; 41% NO) and whether FMSs performance
has met their expectations (35% YES; 25% NO)
.
20% of the FMSs that were installed in the 1990s failed, and were not used in
2001/2002
.
70% of FMS users plan to purchase additional FMS
.
The FMS experience in the United States is consistent with that in Europe
.
RMS configuration (Figure 6.15) is the most popular configuration being used
.
There are no inherent problems in product quality when using FMSs
.
The industry is very dissatisfied with the large initial capital investment in FMSs
.
Industry is not pleased with the high cost of maintenance of FMS
.
In the United States, the actual system productive uptime is 25% lower than
expected
.
90% of all respondents assumed an FMS expected lifetime of 14–15 years in
their economic analy sis
.
Industry is installing systems with more capacity and functionality than
initially needed; this extra capacity was never utilized in 20% of these cases,
and the extra functionality was never utilized in 30% of these cases
172 TRADITIONAL MANUFACTURING SYSTEMS
A student in our class with substantial experience in operating manufacturing
systems expressed his thoughts about the survey:
“The survey of manufacturing industry professionals gave valuable insight into the
thoughts and practices of flexible manufacturing systems. The most surprising finding on
the survey is that manufacturers typically purchase additional machine functionality that is
usually not used. I can relate to this since the manufacturing line that I work with was
purchased with the ability to manufacture V6 and V8 crankshafts. The manufacturing line
has never machined a V6 crankshaft in its 10-year life.”
—Andy Jacobson
PROBLEMS
6.1 Rank the convertibility of the five systems in Figure 6.11.
6.2 The rate at which parts are assembled in station Mb in Figure 6.13 is T seconds,
and in stations Mci is 3T secon ds. Buffer Bb has 10 parts. Station Mc2 stopped
to operate for a period of 9T. How many parts are in the buffer when this station
starts to operate?
6.3 The automated kitting line in Figure 6.14 is adding a tray to the assembly buffer
every 10 seconds. The manual assembly time in the four parallel stations may
vary between 30 and 60 seconds, depending upon the number of components
being assembled. The buffer contained 10 trays when the following assembly
cycles started:
Station 1 30 40 50 60
Station 2 50 30 40 60
Station 3 40 60 40 40
Station 4 60 50 30 50
How many trays are in the buffer after 130 seconds, and how many after
170 seconds?
6.4 A part (cylinder head) has to be machined before manual assembly (see
Figure 6.16). The machining and assembly operations are not balanced:
machining takes 60 seconds per part and assembly takes 40 seconds per part.
Therefore, the machining system works 12 hours a day, and the assembly line
8 hours per day. During the 4 hours in which assembly is not operational, the
machined parts are stored in a buffer and later used by the assembly when
needed. The buffer can store machined parts also during the 8 hours in which
assembly is operational.
PROBLEMS 173
(a) How many parts are machined per day (assume no machine failures), and
how many parts can be assembled per day?
(b) Explain the steady-state operation of the factory (i.e., flow of parts) during
6 minutes of normal production of machining and assembly.
(c) What is the needed buffer maximum capacity?
(d) What happens if the transfer mechanism (e.g., a robot) between the buffer
and the assembly line ceases operation for 30 minutes?
(e) What happens if the machining system stops for 30 minutes during the
operation time of the assembly line?
REFERENCES
1. World Appliance Companies, 4th edn. Appliance Magazine and Industry Statistics, Ltd.,
July 2006.
2. T. Tolio and A. Matta. A method for performance evaluation of automated flow lines. CIRP
Annals, 1998, Vol. 47/1.
3. Y. Koren, F. Jovane, U. Heisel, T. Moriwaki, G. Pritschow, G. Ulsoy, and H. VanBrussel.
Reconfigurable manufacturing systems. CIRP Annals, November 1999, Vol. 48, No. 2,
pp. 6–12.
4. Y. Koren. Computer Control of Manufacturing Systems. McGraw Hill, New York, 1983.
5. J. Browne, D. Dubois, K. Rathmill, S. P. Sethi, and K. E.Stecke. Classification of flexible
manufacturing systems. FMS Magazine, 1984, Vol. 2, No. 2.
6. A. K. Sethi and S. P. Sethi. Flexibility in manufacturing. International Journal of Flexible
Manufacturing, 1990, Vol. 2, pp. 289–326.
7. K. E. Stecke. Formulation and solution of nonlinear integer production planning problems
for flexible manufacturing systems, Management Science, 1983, Vol. 29, No. 3,
pp. 273–288.
8. Y. Koren. Robotics for Engineers. McGraw Hill, New York, 1985.
9. http://www.crux-engineering.com/Automation/Solutions/
10. G. Chryssolouris. Flexibility and its measurement. CIRP Annals, 1996, Vol. 45, No. 2,
pp. 581–587.
Figure 6.16 A factory that includes machining and assembly lines.
174
TRADITIONAL MANUFACTURING SYSTEMS
Chapter 7
Economics of System Design
The cost of building a new manufacturing system may be between $50 million to over
$2 billion (a microprocessor fabrication facility), and its average lifetime is 12–15
years. Such substantial investment that may impact several product generations
requires a rigorous planning that includes a sequence of important decisions:
1. Decide whether to invest at all in a new production system, and, if to invest, in
which type of system .
2. Based on product sale forecasting and estimated capital investment, determine
whether to invest in dedicated, flexible, or portfolio capacity.
3. Calculate the cycle time of each operation and the total time needed for the
whole process to produce one product.
1
4. Optimize the system configuration such that a proper line balancing maximizes
system throughput, and tooling cost is minimized to reduce capital investment
(a methodology to optimize configurations is offered in Chapter 10).
5. Find out the buffer capacity that optimizes the system throughput.
6. Determine the projected operations costs; it is more challenging when flexible
systems that produce several products are employed.
7. Consider system responsiveness to changing orders of customers; responsive-
ness impacts the system throughput.
8. Calculate the optimal speed of each machine; it will impact the whole system
throughput.
The Global Manufacturing Revolution: Product-Process-Business Integration and Reconfigurable Systems
By Yoram Koren
Copyright 2010 John Wiley & Sons, Inc.
175
Designing a system is not a sequential process that follows the above-mentioned
steps. For example, the system configuration affects the tooling cost, which, in turn,
has an effect on the capital cost, but the latter is needed to determine the system type;
the system operations cost impacts profit, and may change decisions about the total
installed capacity. Therefore, the system design is an iterative process that iterates
among the above eight points (and perhaps takes into account more considerations)
until converging gradually to the optimal economic solution.
7.1 LIFE-CYCLE ECONOMICS
Making investment decisions in new manufacturing systems requires knowledge in
engineering as well as in finance and economics. We start, therefore, by defining some
basic economics terms.
7.1.1 Basic Economics Terms
Capital (money) used for building a new manufacturing system should earn returns for
the invested capital. For a new manufacturing system to be economically viable, the
projected sales of the products that the system will produce during its lifeti me must be
higher than the cost of the capital invested in building the system. The cost of the capital
should include the rate of interest paid on the invested capital in building the system.
Return on investment (ROI) is the ratio of money gained (or lost) relative to the
amount of money invested. ROI is given as a percent and usually does not indicate how
long the investment is held. For example, a $10,000 investment that earns $500
generates a 5% ROI. The ROI may be also stated as an annual rate of return.
Cash flow is the balance of the amounts of cash being received by a company
during a given time period. In the context of this chapter, measurement of cash flow
can be used to evaluate the performance of a manufacturing system and validate
whether it generates net income.
Net present value (NPV) is defined as the total present value of an investment
project, found by discounting all present and future receipts and outgoings at an
appropriate rate of interest. If the NPV is positive, it is worthwhile investing in a
project.
2
NPV is an indicator of how much value an investment in a new manufactur-
ing system adds to the firm, and it mea sures the excess or shortfall of cash flows, in
present value (PV) terms.
NPV > 0 means that the investment in a new manufacturing system would add
value to the firm. It is worthwhile to build the system.
NPV < 0 means that the investment in a new manufacturing system would create
loss to the firm, and the project should be rejected.
Example 7.1 A corporation mus t decide whether to introduce a new production
line. The new product will have startup costs, operational costs, and incoming cash
176 ECONOMICS OF SYSTEM DESIGN
flows from selling the manufactured products over 6 years. This project will have an
immediate (t ¼ 0) cash outflow of $100,000 (which includes machinery and employee
training costs). Other cash outflows for years 1–6 are the operation costs that are
expected to be $5,000 per year. Cash inflows are expected to be $30,000 per year for
years 1–6 and there are no cash flows expected after year 6. The interest rate is 10%.
The PV can be calculated for each year:
T ¼ 0 : $100; 000=1:10
0
¼$100; 000 PV
T ¼ 1 : ð$30; 000$5000Þ=1:10
1
¼ $22; 727 PV
T ¼ 2 : ð$30; 000$5000Þ=1:10
2
¼ $20; 661 PV
T ¼ 3 : ð$30; 000$5000Þ=1:10
3
¼ $18; 783 PV
T ¼ 4 : ð$30; 000$5000Þ=1:10
4
¼ $17; 075 PV
T ¼ 5 : ð$30; 000$5000Þ=1:10
5
¼ $15; 523 PV
T ¼ 6 : ð$30; 000$5000Þ=1:10
6
¼ $14; 112 PV
The sum of all these PVs is the NPV, which equals $8881. Since the NPV is greater
than zero, the corpo ration will have a profit by investing in the new product line.
7.1.2 Life-cycle Econo mics of Manufacturing Syste ms
The dedicated manufacturing line (DML) is an ideal solution when a part design is
fixed, the change in its design is not imminent, and demand is fixed. A Flexible
manufacturing system, although more expensive than DML, enables the production
of new members of the same part family on an existing system, which, in turn, expands
the lifetime of the system.
In the example below, we use NPV to evaluate the life-cycle investments in DML
and FMS when producing a single product. By using a simple economic model, we
study the effects of product life cycle and equipment reusability of each production
system on the investment strategy of a manufacturing firm. The results illuminate
some basic concerns regarding the fast-changing products and the need to reconfigure
the production line frequently.
Example 7.2
3
Consider an engine block manufacturer planning to fulfill demand
of three (3) product generations over the next 12 years where each product
generation lasts 4 years. The demand volume is assumed to be 300,000 engine
blocks per year, and the profit per part is $120. The manufacturer may invest in either
DML or FMS. The DML capital cost is $100 million, and that of FMS is $130
million. Once the next product generation arrives the same FMS can be reutilized,
but there is a need for new fixtures, new tools, and ramp-up that cost $25 million. At
the end of its lifetime (12 years), the salvage value of the FMS (selling used CNC
machining centers) is $17 million. Each time a new product generation is manu-
factured a new DML must be built, but certain parts of the existing DML, which is
worth $20 million, can be either reutilized for the new line or sold (have a salvage
value). The annual interest rate is 10%.
LIFE-CYCLE ECONOMICS 177
Table 7.1 shows the cash flow forecast and PV of using DML and FMS. Since the
FMS NPV is greater than the DML NPV, building the FMS is more economical.
Now let us examine the case for different product lifetimes—of 2 years and of 6
years. The results are shown in Tables 7.2 and 7.3, accordingly. We see that if the
expected product lifetime is 6 years, the two systems create equal values.
A similar calculation was done for a product life cycle of 1 year, which yields an
NPV of $42 million for FMS, and $367 million for DML.
Figure 7.1 shows the NPV comparison between the two systems, with the
following conclusions (for the data given above):
1. If the expected product lifetime is less than 1.5 years, building any system is a
loss.
2. If the expected product lifetime is more than 6 years, install a dedicated
machining line.
3. If the expected product lifetime is between 2 and 6 years, install the FMS.
Although these conclusions are drawn for a particular example, they may be
considered as general rules for system life-cycle economic planning.
7.2 CAPACITY PLANNING STRATEGIES
When planning a new manufacturing system for producing several products, the
optimal investment in the system capacity is a major deci sion to make. This decision,
which has long-term economic consequences, addresses two major issues: (1) How
much capacity to build? (2) Whether to invest in dedicated or flexible resources, or a
portfolio consisting of both dedicated and flexible systems. Flexible systems possess
the ability to produce multiple types of products very economically and quickly.
Therefore, by switchi ng from one p roduct to another, they may alleviate unfavorable
effects of demand uncertainties. However, the versatility to produce multiple products
Figure 7.1 An economic comparison of two systems: dedicated and FMS.
178
ECONOMICS OF SYSTEM DESIGN
TABLE 7.1 Cash Flow (in $ million) and Net Present Value for DML and FMS for Product Life Cycle of 4 Years
Year 0 1 2 3 4 5 6 7 8 9 10 11 12
DML Cash flow 100 36 36 36 36 36 36 36 36 36 36 36 36 NPV
þ 20 þ 20 þ 20
100 100
PV 100 32.7 29.8 27 30 22.3 20.3 18.5 20.5 15.3 13.9 12.6 17.8 59.7
FMS Cash flow 130 36 36 36 36 36 36 36 36 36 36 36 36 NPV
25 25 þ 17
PV 130 32.7 29.8 27 7.5 22.3 20.3 18.5 5.1 15.3 13.9 12.6 16.9 92
DML(NPV) ¼ $59.7 million FMS(NPV) ¼ $92.0 million.
TABLE 7.2 NPV for Product Life Cycle of 2 Years
Year 0 1 2 3 4 5 6 7 8 9 10 11 12 NPV
DML PV 100 32.7 36.4 27 30 22.4 24.8 18.5 20.5 15.3 17 12.6 17.8 82.4
FMS PV 130 32.7 9.1 27 7.5 22.4 6.2 18.5 5.1 15.3 4.2 12.6 16.9 47.5
TABLE 7.3 NPV for Product Life Cycle of 6 Years
Year 0 1 2 3 4 5 6 7 8 9 10 11 12 NPV
DML PV 100 32.7 29.8 27 24.6 22.4 24.9 18.5 16.8 15.3 13.9 12.6 17.8 106.5
FMS PV 130 32.7 29.7 27 24.6 22.3 6.2 18.5 16.8 15.3 13.9 12.6 16.9 106.5
179
often requires higher investment costs compared to dedicated systems that can only
produce one type of product.
In most industries, capacity investment decisions are made before demand is
observed and the optimal capacity choices may vary from one firm to another even in
the same industry. The investment decision on the amount and type of capacity—
dedicated, flexible, or a portfolio of dedicated and flexible systems—is mainly
influenced by the following factors:
1. The number of products to be produced simultaneously in the plan t.
2. Investment costs in dedicated versus flexible systems.
3. Product marginal revenues.
4. Product demand volatility during the planning horizon.
5. The frequency of product changes and the expected lifetime of products.
The Number of Products to be Produced Simultaneously: Flexible technology
can deal with changes or uncertainty in demand mix. It enables to change the mix of
products manufactured in a plant, and produce more of highly profitable products
when their demand surges. Usually, if a plant produces more than four, five products
simultaneously, the decision will be to invest in flexible capacity.
Investment Costs of Dedicated Versus Flexible Systems: When producing large
quantities, the investment cost in flexible systems is always larger than that in
dedicated lines. The margin may be 10–100% in large machining systems. Flexible
capacities are usually favored more when their investment costs are closer to those of
dedicated lines. When the manufactured quantities are small, dedicated lines are not
cheaper than flexible systems, and the latter is the optimal solution.
Product Revenues: Higher profit margins and higher prices of the product
produced warrant a higher investment level, since losing sal es of products causes a
significant financial loss. Installing flexible capacity in such cases is economically
justified.
Product Demand Volatility During the Planning Horizon: Investment in
flexible capacity hedges against uncertainty in future demand, since production can
be easily shifted from a product with diminishing demand to a product with rising
demand. Therefore, when market volatilities are high, an investment in flexible
capacity is the right economic choice.
The Frequency of Product Changes and the Expected Lifetime of Products:
When a firm plans to rapidly introduce new product models in the near future and
expanding its product scope, the firm should invest in flexible capacity.
When the above-mentioned considerations are not conclusive beyond all doubt, a
firm may invest in a portfolio consisting of both dedicated and flexible systems.
To gain further insights, we provide below a formulation for the optimal capacity
investment problem and demonstrate a solution for numerical examples. The
examples show at what circumstances profits are maximized when considering three
investment strategies: dedicated capacity only, flexible capacity only, and a portfolio.
180 ECONOMICS OF SYSTEM DESIGN