Koren Y. The Global Manufacturing Revolution: Product-Process-Business Integration and Reconfigurable Systems

Подождите немного. Документ загружается.

(b) What is the recommended cutting speed for the ﬁrst machine in this two-

machine system?

The speed of ﬁrst machine should produce parts at the similar rate of the

second machine. In an 8-hour shift, the second machine produces

8  60

6:5



¼ 73 parts.

From the answer of question a, we see that at a cutting speed 300 m/minute, the

number of parts produced on the ﬁrst machine, 74, is the closest to 73.

Therefore, the recommended cutting speed on the ﬁrst machine for two-

machine operation is 300 m/minute. In this case the system will produce 73

parts per shift, and only seven tools will have to be changed during the shift.

Conclusion: The optimal cutting speed on each machine in a system (or the operation

speed in assembly systems) depends on the system conﬁguration. The goal is to

maximize the whole system throughput, rather than the individual machine

throughput.

PROBLEMS

7.1 AVP of a manufacturing ﬁrm that produces two products has to decide (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.

The VP received the following forecasting for the demand of each product.

Year 1–2 Year 3–6

Volume Volume Probability

Product A 500,000 200,000 0.7

600,000 0.3

Product B 200,000 200,000 0.3

600,000 0.7

A dedicated line can produce 300,000 annually and its cost is $30 million.

Flexible systems are supplied at chunks; each has an annual production capacity

of 50,000 units and costs $7 million. The revenues from each product sold are

$40.

(a) Calculate the optimal capacity—type and quantity.

(b) What is the optimal solution if the cost of a dedicated line that can produce

200,000 annually is $15 million?

300,000 annually is $40 million?

PROBLEMS 201

7.2 A factory manufactures two products A and B on a ﬂexible system during

15 hours every day for 5 days per week. The cycle time of part A is 9 minutes,

and that of B is 12 minutes. The switchover time from one product to the other is

20 minutes.

The customer requires shipping equal quantities of A and B once a day in

containers of 10 parts each (separate containers for part A and for B). Only full

containers with 10 parts are being shipped. A few parts may be stored overnight

for next day delivery.

(a) How many parts are shipped to the customer every week if the system

operates with n switchovers per day, where n ¼ 1, 2, 3, 4, and 5. Calculate the

lot sizes in each case.

(b) If the customer requires shipping parts twice a day (not neces sarily at certain

times), what is the recommended n?

containers of A and B every 4 hours. What is the recommended number

of daily switchovers (n) to satisfy this requirement?

7.3 Calculate the total number of tools in the tool magazines of all machines in the

three systems depicted in Figure 7.9, if each machine is capable of producing

both part A and part B. Which of three systems in Figure 7.9 is the most

economical in your opinion?

7.4 A part has to be machined on a single machine tool and you have to recommend a

cutting speed. If the cutting speed is higher, more parts are machined, but the

tool life deteriorates exponentially, as shown in the table below:

Cutting speed (m/minute) 150 240 300 480 600

Tool life (minute) 240 80 60 20 15

Time to machine one part 16 10 8 5 4

The tool-changing time of worn tools is on average 8 minutes.

The plant is working in two shifts, each of 8 hours, with 1-hour break between

shifts.

(a) Calculate the number of parts machined in 8 hours at each of the above

speeds (reme mber to include the tool changing time in your calculations).

(b) Calculate the number of tools needed in 8 hours at each speed.

recommended cutting speed?

7.5 Assuming that two machines are needed to complete a part, a robot transfers

the semi-machined part from one machine to the next in 30 seconds

(see Figure 7.13). If identical cutting speeds are used on both machines, the

202 ECONOMICS OF SYSTEM DESIGN

cutting time per part on Machine 1 is 40 seconds and the cutting time on

Machine 2 is 50 seconds.

(a) What is the throughput of the line (assuming machine reliability 100%) in

parts per 8-hour shift?

(b) Can you recommend a change (in %) in the cutting speed of the machines in

order to increase the system throughput?

REFERENCES

1. The cycle time is calculated with the aid of process plans. A process plan determines the

sequence of individual manufacturing operations needed to produce a given part or product,

in which the machining parameters for each opearion are given. We do not discuss this topic

in the book. For reference you may see: W. Wencai and D. Yip-Hoi: Modeling and

identiﬁcation of feed drive kinematics and cycle time calculation. ASME International

Mechanical Engineering Congress, Washington, DC, 2003.

2. J. Black. Oxford Dictionary of Economics . Oxford University Press, 2003.

3. R. Katz, I. Duenyas, W. Norongwanich, and Y. Koren. Economic beneﬁts of reconﬁgurable

manufacturing systems (RMS). Global Powertrain Conference, Ann Arbor, MI, September

2002.

4. Y. Koren, F. Jovane, U. Heisel, T. Moriwaki, G. Pritschow, AG. Ulsoy, and H. VanBrussel.

Reconﬁgurable manufacturing systems. Annals of the CIRP, 1999, Vol. 48, No. 2,

pp. 6–12.

5. J. A. Van Mieghem. Investment strategies for ﬂexible resources. Management Science,

1998, Vol. 44, No. 8, pp. 1071–1078.

6. O. Ceryan and Y. Koren. Manufacturing capacity planning strategies. Annals of the CIRP,

August 2009, Vol. 58, No. 1.

7. P. Damodaran, K. Srihari, and S. Lam. Scheduling a capacitated batch-processing machine

to minimize makespan. Robotics and Computer-Integrated Manufacturing , April 2007, Vol.

23, No. 2, pp. 208–216.

Figure 7.13 A machining system of two machines.

REFERENCES 203

8. J. B. Matson and D. McFarlane. Assessing the responsiveness of existing production

operations. International Journal of Operations and Production Management, 1999, Vol.

19, No. 8, pp. 765–784.

9. R. Shafaei and P. Brunn. Workshop scheduling using practical (inaccurate) data—part 1: the

performance of heuristic scheduling rules in a dynamic job shop environment using a rolling

time horizon approach. International Journal of Production Research, 1999, Vol. 37, No.

17, pp. 3913–3925.

204

ECONOMICS OF SYSTEM DESIGN

Chapter 8

Reconﬁgurable Machines

To illustrate the concept of reconﬁgurable machines (RMs) let us look at mechanical

wrenches (Figure 8 .1). An open-end wrench that is dedicated to grip just one-size bolt

(ﬁxed part geometry), the adjustable wrench that is a ﬂexible tool (ﬁts any part

geometry), and the reconﬁgurable, modular socket wrench, which can turn bolts

whose geometry ﬁts a bolt family. The perfect ﬁt between the socket and the

part geometry enables quicker working speeds, and the changing sockets enable

a functionality ﬁt to a whole part family of bolts.

The modular tool with its sockets represents one type of reconﬁgurability that

depends on modularity. But modularity is not a necessary condition for reconﬁgur-

ability. A structure may be reconﬁgurable without being modular as the following

example illustrates.

The Global Manufacturing Revolution: Product-Process-Business Integration and Reconﬁgurable Systems

By Yoram Koren

Figure 8.1 Reconﬁgurable tools ﬁll the gap between dedicated tools and ﬂexible tools.

205

Illustration by R. Hill

8.1 THE RATI ONALE FOR RECONFIGURABLE MACHINES

A RM is a machine whose structures can be altered to provide either alternative

functionality or incremental increase in its production rate in order to meet changing

demand. The RM can be continuously returned to its original state, or modiﬁed to

provide new functionality or production capacity as needed. RM design has,

therefore, two basic objectives:

1. To adapt the machine functionality to ﬁt a new member of a family of parts

(...exactly the functionality needed)

2. To increase the machine production rate by adding resources

(...exactly the capacity needed)

RMs are designed around the common characteristics of part families, and this

feature differentiates them from dedicated or ﬂexible machines. For example, while

there might be major differences between various engine cylinder heads, the basic

conﬁguration of all of them is quite simi lar, as depicted in Figure 8.2. A reconﬁgurable

machine tool (RMT) can therefore be designed to perform the necessary machining

operations common to all the members of the part family with reconﬁguration to the

machine itself. Similarly, a reconﬁgurable inspection machine (RIM) can be designed

to perform the appropriate inspections on a set of features for members of the part

family and then reconﬁgure to inspect another set of features. The RM concept can

cost-effectively produce or inspect a whole family of parts, even part styles that have

not yet been called for.

An excellent demonstration of what we

mean by reconﬁgurable structures are toy

action ﬁgures that are transf ormable (from

a robot to a truck, for example). The capa-

bility of transformation is always present in

the toy whatever conﬁguration is desi red at

the time. By simply applying a series of

manipulations, the child modiﬁes the toy

from one form to the other without adding

or subtracting anything. Likewise, with one

class of non-modular reco nﬁgurable ma-

chines, just modifying the structure can

change functionality without additional

parts.

206 RECONFIGURABLE MACHINES

RMs constitute a new class of production and inspection machines that ﬁll the gap

between dedicated mac hines and ﬂexible machines, as shown in Table 8.1

The RM sacriﬁces some general ﬂexibility for faster throughput speed in perform-

ing the task. To create an RM, the part family (including all potential members) must

be understood before the machine is designed. This allows the RM to be designed for

repetitive, fast operation like dedicated tools. But unlike dedicated tools, RMs have

features that allow enough ﬂexibility (extending the envelop reach, changing whole

tool magazines, adding functions) that they can produce any of the part family

members, in quantity and on demand. RM performance in an environment of volatile

markets makes them superior to ﬂexible machines that are, by contras t, built with

the widest range of ﬂexibility possible, (like coordinate measuring machine, CMM),

and only then the application is made suitable for a speciﬁc part (usually by

programming).

The RM paradigm is driven primarily by economic considerations. Rather than

custom-build a multi-spindle dedicated machine, for example, a manufacturer could

purchase a single-spi ndle reconﬁgurable machine and then add spindles to it; allowing

it to cut several parts at the same time as market demand justiﬁed the investment

(an adju stable-capacity RM). Also, rather than invest in a highly complex, general-

purpose computerized numerical control (CNC) machine tool, it would be more

economical to have a simpler machine with enough functionality to produce a whole

Figure 8.2 A part family of cylinder heads: the basic machining features are similar.

TABLE 8.1 Machine Types

Machine Type Dedicated Reconﬁgurable Flexible

Given part geometry Fixed part

geometry

Part geometry ﬁts

a part family

Any part geometry

Operation speed Very fast Fast Slow

Flexibility Not ﬂexible Customized ﬂexibility

for a part family

Full ﬂexibility

THE RATIONALE FOR RECONFIGURABLE MACHINES 207

family of parts without buying a lot of extra unused functionality (an adjustable

functionality RM). Focusing one’s purchase on only the capacity and functionality

needed for a desired part family gives the RM an advantage.

The primary aim of a RM is to cope with changes in the products or parts to be

manufactured. The possible changes that are considered:

Part geometry

Production volume and production rate

Required processes

Material property, such as kind of material, hardness, etc.

8.1.1 Convertibility—Fitting Functionality to the Part Geometry

Adding new motion units to increase the number of axes-of-motion, or changing out

one unit for another having a different degree of freedom (DOF), may change machine

functionality for different part geometries in the same family. Figure 8.3 shows

a schematic of a possible reconﬁgurable ﬁve-axis machining center that includes two

optional rotary axes for rotational motions.

8.1.2 Scalability—Changing Production Volume and Rate

In order to increase the rate of production, the capacity of a machine spindle unit can

be increased by changing from a single-, dual-, or even multi-spindle unit. The multi-

Figure 8.3 Components for a reconﬁgurable rotary axis.

208

RECONFIGURABLE MACHINES

spindle unit is a very powerful tool to increase productivity, performi ng multiple

operations simultaneously. Modularized spindle units capable of different speed

ranges and horsepower are another good use of RMTs. Figure 8.4 shows an example of

a RMT where a part mounted on a rotary table is being machined simultaneously by

four spindles.

The number of spindles in the example of Figure 8.4 can be varied to accommodate

the desired production rate. Each spindle is a Z-axis module. Two additional manual

axes are the location of the spindle and its cutting angle.

8.1.3 Reconﬁgurability for Changes in the Machining Process

More than just the cutting tool, sometimes the spindle type, or even the machine tool

conﬁguration can be changed to cope with changes in the machining process. In some

applications turning can be performed not only on a turning center but also on

a milling machine, and milling and drilling operations can be performed on a lathe

by using a milling spindle that replaces the ﬁxed tool post. Another example is the

machine in Figure 8.4 that can be converted to a vertical turning center, in which parts

are machin ed simultaneously with multiple tools for drilling, and turning operations.

8.1.4 Reconﬁgurability for Material Properties

Milling titanium, for example, requires high torque at low speed, while milling

aluminum requires the opposite—low torque at high speed. That is why there are

milling machines in the aerospace industry that were designed for changeable

spindles. Spindles are rapidly changed according to the workpiece material.

Figure 8.4 Top view of a vertical turning–milling center with multiple spindles.

THE RATIONALE FOR RECONFIGURABLE MACHINES 209

8.2 CHARACTERISTICS AND PRINCIPLES OF RECONFIGURABLE

MACHINES

A RM is designed around a deﬁned part family. Approaching machine design from

this point of view creates a less complex, albeit less ﬂexible machine. Still it is

a machine that contains all the functionality and ﬂexibility needed to produce (or

inspect) a whole part family. This characteristic is called Customization or Custom-

ized Flexibility; it is in contrast to the general ﬂexibility characteristic of a CNC or

a CMM that is commonly used in inspection.

We deﬁne a RM as follows:

A RM is a machine designed for rapid change in its structure in order to quickly

adjust production capacity and/or functionality within a part family.

RMs can be easily switched from handling (i.e., machining, assembly, or inspec-

tion) one part of the family to another part. This is called machine responsiveness. To

enhance the RM responsiveness, there is a set of core characteristics that should be

embedded in the machine system, both in the mechanical structure and in its controls.

In general, RMs possess the six core characteristics that are deﬁned below.

Customization The ability to apply a customized ﬂexibility to production

or inspection machines to meet new requirements

within a part family.

Scalability The ability to efﬁciently change the machine’s production

throughput by altering or augmenting the components

in the machine.

Convertibility The ability to efﬁciently redirect the functionality of the

machine and its controls to suit new production

requirements.

Modularity The compartmentalization of operational functions and

hardware into units that can be manipulated between

alternate machine conﬁgurations.

Integrability The ability to integrate machine modules rapidly and

precisely by a set of mechanical, informational,

and control interfaces.

Diagnosability The capability of monitoring the current state of

a machine and controls so as to detect and diagnose

the root cause of output product defects.

The ﬁrst three characteristics: customization, scalability, and convertibility are

critical to creating cost-effective machin es. An RM with customized ﬂexibility will be

less expensive to build and operate than a comparable CNC machine that has general

210 RECONFIGURABLE MACHINES