Koren Y. The Global Manufacturing Revolution: Product-Process-Business Integration and Reconfigurable Systems
Подождите немного. Документ загружается.
12.2.2 Information-aided Employee Empowerment
Jack Welch (the former CEO of General Electric ) was a prominent advocate of
employee empowerment in the 1990s. In a speech in 1992 he said: “The way to
harness the power of the employees is to protect them, not to sit on them, turn them
loose, let them go—get the management layers off their backs, and the functional
barriers out of their way.”
2
In a 1993 interview with USA Today, Welch said:
*
“We
generally used to tell people what to do, and they did exactly what they were told to do.
Now this task (of improving productivity) is to be shared with the men and women on
the factory floor, and we are constantly amazed by how much people will do when they
are not told what to do by the management.” In 1994 he said: “We want everyone to
have a say. We want ideas from everyone.”
*
The Lean Production principle of pushing responsibilities down the organizational
ladder and empowering workers on the production floor is another important step in
this bottom-up approach. In lean production each team has quality checking respon-
sibilities and can stop the entire production line when a problem appears.
However, Mr. Welch’s empowerment philosophy does not emphasize the role of
information technology (IT) in empowering the workforce. Applying IT to empower
people requires a shift from labor-intensive to brain-intensive operations even at lower
levels of the hierarchy, a shift that will affect the skills that both engineers and workers
must possess.
In the global manufacturing revolution, employee empowerment is a competitive
necessity. But to make the right decisions, the employee must have the right
information, and be capab le of understanding it. The day is not so far off when all
factory employees will have to hold at least a college Associate Degree to do their
jobs. Information-aided empowerment of employees, combined with raising their
education level, will be critical to increase productivity and profits for factories in the
Western world.
Enterprise strategists will also need to contribute to the information-aided orga-
nizational structure by looking at issues such as:
.
What kind of information is needed to make more effective decisions at various
levels of the organization?
.
How should the plant manager and the enterprise best reconfigure resources in
order to enhance productivity?
.
How can information and manufacturing processes support corporation
strategy?
This empowering bottom-up approach should also be implemented in the higher
levels of the industry hierarchy. Enterprise leaders make intelligent decisions and
make them rapidly. This bottom-up, information-rich approach, with technology
aiding in empowering people at all levels, is a radical change that will create a less
*
Quoted in USA Today, July 26, 1993.
*Quoted in Industry Week, May 2, 1994.
TWENTY-FIRST CENTURY IT-BASED ORGANIZATIONAL STRUCTURE 321
corporately conservative industry. It will create an industry that has an entrepreneurial
spirit embedded in its structure, a spirit that will be needed to compete in the global
environment of the twenty-first century.
The new requirements from the workforce will change the workers skill set. In a
speech in Montreal (2003) Dr. Ezio Andreta, Director, Industrial Technologies
Research at the European Commission, said: “The next industrial era will be
characterized by a new economy based on knowledge. The shift from labor-intensive
to brain-intensive operations will modify the jobs and skills required of engineers and
the entire wor kforce.”
12.2.3 Organizational Structure for Developing New Products
Whether instructions flow top-down from the executive at each layer of the orga-
nization, or from the bottom-up, and employees are empowered, the organizational
structure is still a vertically integrated, pyramid-type organization (Figure 12.1 shows
examples). There are, however, other organizational structures possible. Two dis-
tinctly different organizational structures, currently used for new product develop-
ment, are the matrix organization and the project-based organization.
3
Figure 12.2 describes an example of a matrix organization for a manufacturing
company that is continuously developing new products. The project leaders and the
functional departments have more or less equal power, but the ultimate responsibility
to deliver the project’s expected deliverables is on the project leader. This structure
combines functional skills with resources directed toward common goals more
efficiently.
Figure 12.2 A matrix organizational structure.
322
IT-BASED ENTERPRISE ORGANIZATIONAL STRUCTURE
The project-based organization is commonly found in the movie industry. For each
new movie, actors and actresses are selected according to their fit with the movie and
its budget. When the project ends, the team is dissolved. This approach is also
frequently implemented in the development of new products in large manufacturing
companies. For example, people from different functional departments of the
organization are brought together into a team to develop a conceptual design for a
new product (e.g., a new vehicle). Each person’s specialty is in a different area, and the
group dynamics and team interactions enable cross-fertilization of ideas and often
result in innovative products. Members return to their previous departmental post
when the team is dissolved.
12.3 INFORMATION TRANSFER IN MANUFACTURING SYSTEMS
As we have said, ente rprise control and management should be modeled along
the lines that the Army has demonstrated in recent operations. This premise is
explained below.
12.3.1 Control and Management Levels
A typical enterpris e for producing discrete products is composed of several levels.
At the bottom level is the individual piece of equipment, such as a machine tool, a
robot, a part-handling gantry, a fixture to carry the parts, a buffer to hold parts
before processing, a rotary index table to assist in part loading into the machine, an
ambient temperature sensor, and perhaps an AGV or some other transport device.
Each of these pieces of equipment has its own controller; we call it a machine-level
controller.
The next level in the hierarchy is the cell level. A cell contains a group of machines
that are doing a particular operation as well as all the auxiliary equipment that is
needed to complete the operation. An example of a cell may be several robots doing
simultaneous welding opera tions on the same auto-body in an automotive assembly
line. Another example may be a group of machine tools in which the part to
be machined is transferred from one machine to the next until completion of an
operation phase.
4
An additional example of a cell may be a group of identical machine tools that are
machining similar parts in parallel, as well as all their part-handling equipment, as
depicted in Figure 12.3. The part-handling equipment may include a buffer to hold
parts before proce ssing in the cell, a gantry equipped with grippers to bring parts from
the buffer to an indexing table from which parts are loaded into the machine, and the
indexing table itself. All these pieces of equipment must work in harmony under the
cell controller coordinati on program.
Next, the cell-level controllers are connected via a network to the system-level
controller, as depicted in Figure 12.3. A factory facility may include several
manufacturing systems. For example, Ford Motor Co.’s Windsor engine plant
INFORMATION TRANSFER IN MANUFACTURING SYSTEMS 323
contains four RMS machining systems to produce engine parts, with a total of 180
CNC machining centers. (The equivalent in the Army might be a brigade of 180
tanks.) Such a factory may include more than one overall manufacturing activity, say,
a machining system that completes final finishing on some parts and an assembly
system that combines the parts from the machining system and other sources and
assembles them into complete products. A factory may include many combina tions of
systems, all in one facility.
In our twenty-first-century IT-support model, the enterprise system controller is at
the top. It accumulates the customers’ orders and coordinate the supply of outsource
parts from the various suppliers. It also sends daily product demand requests to all the
appropriate factories and manufacturing systems. The customers are at the very top
Figure 12.3 Enterprise hierarchical control levels.
324
IT-BASED ENTERPRISE ORGANIZATIONAL STRUCTURE
and the entire enterprise structure is built in order to benefit them—providing them
with the right product in a timely manner.
We envision that enterprise control will move in the same dir ection of bottom-up
complex information systems that the Army has recently adopted. Similar to the
Army’s modern tank that sends information about the battleground, the machine M
ij
in the factory will send information about its conditions to the cell controller and on
through the system.
One type of information that a machine-level controller might provide is
machine temperature. If the temperature exceeds a certain level, the quality of the
part may deteriorate because the machine loses accuracy. For instance, under
elevated heating conditions, dimensions expand causing the machine to lose the
calibration that it held at normal room temperature. But the machine should not
make the decision to stop production by itself. The degree of critical precision
should be determined at a higher level of hierarchy (probably at the system level),
and that level then communicates instructions as the temperature approaches the
critical threshold.
To be effective, however, the machine controller should not send raw data, but
convert the data into coherent information before it is sent. In this example, the
machine should not send its continuous temperature measurements to the cell
controller; it should send just a warning signal when the temperature crosses a
certain threshold. The cell controller should monitor all machines in the cell and
whether any others have crossed the temperature threshold, and then send appropriate
signals to the system-level controller. The latter will make the decision whe ther to stop
production. This decision may be either automatic or done by a person (in the example
in Figure 12.3—it is a director’s decision).
12.3.2 Coordinating Production
Establishing factory-level IT is essential in coordinating production of several
systems. Figure 12.4 shows an example in which engine blocks are sent from casting
to a large machining system that machines both V6 and V8 blocks; the machined
blocks are then sent to engine assembly. There are two assembly systems, one for V6
engines and the other for V8 engines. When either of these systems has a fatal failure
and stops operating, there is no need to continue to supply engine blocks to that
system. It is the task of the factor y-level IT to coordinate the production of these
Casting
Machining System
V6 and V8
Assembly V6
Assembly V8
Information
Information
Figure 12.4 Several manufacturing systems are linked in order to produce products.
INFORMATION TRANSFER IN MANUFACTURING SYSTEMS 325
systems in this case and the upstream and downstream operations in the general case.
The shutdown in the asse mbly system in this example may also be utilized for
scheduled maintenance.
12.3.3 Real Experience in Industrial Plants
From the above discussion, one may draw the following conclusions:
.
Factory IT should be utilized to empower workers at all levels.
.
IT enables blending top-down strategic guidance with bottom-up tactical
decisions.
.
IT should be embedded into all levels of the enterprise, from the machine
controller, through the cell controller, to the factory controller, and above.
Empowering factory employees with timely, relevant information is essential to
highly responsive operations and enterprise success. We asked some knowledgeable
readers of this book about their experience on the factory floor. Here is a small
collection of real-life stories.
“In my company I have definitely seen the rigid top down structure. The motto of
our previous Area Manager was ‘the easiest thing to do is follow instructions. That
motto led to a group of disenfranchised, unhappy workers that would willingly
carry out an incorrect order because we had been beaten down to accept things as
they were told to us. I have always felt the need for the bottom-up information
propagation and delegation authority at lower levels, but I think that our depart-
ment has a long road to travel to get out of the old way of thinking.” BJ Baker
“Employees must be com fortable about confronting their superiors. Too often,
employees don’t feel that they can give honest opinions to their supervisors, but
this is needed in order for change to occur. The people who are most attuned to
which direction a company should move are frequently the people at the bottom.
Executives are too far removed from the action and the products to know what
should happen day-to-day.” Andrea Thomas
“At the plant where I work, we have the technol ogy and the facility to operate as
bottom-up organization, but the top-down style is so prevalent in the minds of the
old-fashioned management leaders that I wonder if the transition to bottom up will
ever actually happen.” Andres Babian
“A union environment, such as what is present (2004) in most automotive assembly
plants, tends to make this philosophy more difficult to imagine. In the plant that I
work in, the union is very particular about making sure that everyone only does the
job that they are assigned to. When people are concerned that a union grievance
may be written against them, they may be less likely to go beyond the scope of their
job. It is possible for someone who finds a quality defect to have a grievance written
against him or her if they point it out because their job is not a quality auditor. This
philosophy goes against the very idea of empowered employees.” Cindie Nieman
326 IT-BASED ENTERPRISE ORGANIZATIONAL STRUCTURE
12.4 IT-BASED MAINTENANCE OF LARGE SYSTEMS
This textbook has already described how large manufacturing systems in factories
contain many production machines, material handling, and other pieces of equip-
ment—all of which may break down during normal operation. These systems require
(1) regularly scheduled maintenance, (2) repairs of machines that break down and
stop operating, and (3) incidental maintenance tasks that, although urgent, require
relatively small effort, such as adding coolant or replacing tools. Appropriately
coordinated maintenance-scheduling decisions can increase the system productivity
if they are done based on information that is transferred comprehensively across
hierarchical levels of control and management.
Large manufacturing systems typically have specific personnel to perform these
maintenance and repair jobs, each person assigned to a different task depending on
his/her skill level and set (e.g., electrician or mechanics). These specialists are
typically empowered to do all that is required to sustain production: repair, main-
tenance, adding coolant, etc. On the factory floor it is hard to assess the impact of a
breakdown on the factory throughput and determine what to do first. If an unscheduled
event (e.g., a machine breakdown) occurs, or if several events occur simultaneously,
which repair job has the highest priority? Which machine failure most seriously
endangers the production schedule?
Maintenance crews are frequently facing the challenge of determining the priority
of completing their tasks because of the lack of a system atic decision-making system
for maintenance. Without comprehending the whole picture, maintenance personnel
cannot decide whether they should drop their current job and move to a new problem,
or concentrate more resources to speed the completion of the current maintenance job.
A solution to this problem can have significant and ongoing benefits to the operation.
On top of it there are always conflicts between the production manager, who wants to
keep production up no matter what might happen to the production machines, and the
maintenance manager.
“There is often a conflict between the need for production and that for mainte-
nance. Production managers prefer to keep the production system running in order
to meet daily production target, while maintenance managers hope to get some
sufficient machine time so that adequate preventive maintenances can be per-
formed on the equipment. Resolving this conflict requires the development of
advanced decision-making support tools.” Professor Jun Ni.
Decision making for effective maintenance of large systems is complex because it
depends on several independent sourc es of information:
.
Current “health” status of each machine, and each material-handling device,
including: down; running; idling; being maintained; or just about to breakdown
.
The current ongoing repair schedule
IT-BASED MAINTENANCE OF LARGE SYSTEMS 327
.
The scheduled maint enance plan for the day and the next several days
.
The history of each repair, which is a factor in determining a machine’s
reliability
.
The daily production target, and the rate at which it is being accomplished
.
Possibility of re-routing of jobs
.
The standar d system configuration, the given takt time, and optional alternatives.
The system configuration has a profound effect on the maintenance decision-
making process. For example, some configurations have identical machines doing
identical manufacturing tasks in parallel, as shown in Figure 12.3. If one of these
identical machines is already down and in repair, and another of these machines goes
down, then the urgency of repair of at least one of them becomes very high. Or, as
another example, in the configuration of Figure 12.3, if one gantry is down, the whole
system is down, stopping the entire production. In this example, repairing the gantry
becomes the highest priority.
The following types of information should be available and should be integrated
into a multi-level decision-making system in order to make the appropriate priority
choices with the highest impact on productivity.
1. Information from the machine level (Level 1 in Figure 12.3): The “health”
status of the machine, as obtained from sensory data that is available from each
machine. For example:
.
The machine is down and requires repair (e.g., a broken cutting tool o r
spindle).
.
The machine has slightly deteriorated and will need an upcoming mainte-
nance action within a certain time horizon (e.g., to change a worn cutting tool,
or a worn bearing, or motor shaft).
.
The machine has severely deteriorated and requires an urgent repair within a
specified time horizon (e.g., a spindle bearing is wearing out quicker than
expected).
.
The machine is healthy and does not require any repair in the plan ning
horizon.
2. Information from the cell level (Level 2):
.
The machine workload: the production volume of each machine during the
defined time period (e.g., one day or one shift).
.
The time window available for repair or maintenance in any machine in the
cell.
3. Information from the system level (Level 3):
.
The availability of maintenance crews.
.
Spare part inventory.
.
Replacement tools.
.
Routine maintenance scheduling.
328 IT-BASED ENTERPRISE ORGANIZATIONAL STRUCTURE
.
Alternative routing options for the parts produced on the system.
4. Information from the enterprise level (Level 4 or 5):
.
The production requirement (e.g., the mix of products and the number of
products to be produced per shift).
Recent research on maintenance deci sion support tools has made significant
progress. Instead of the heuristic rules that are commonly used in today’s manufactur-
ing plants for the prioritizatio n of maintenance work-order requests, systematic
decisions can be established.
5
These decision-making tools are answering the
question raised above “ which repair job has the highest priority?” by systematically
considering the four types of information listed above (especially the production
demand, system configurations, the criticality of the impending maintenance or repair
requests, and the availability of maintenance crews and their skill sets—electricians,
mechanics). This kind of systematic decision-making tools is especially effective to
deal with the complex interaction between random production equipment failures and
available maintenance resources.
In conventional production system design practice, production bottlenecks are
considered only in a static sense or based on steady-state conditions. However, in
actual production, the short-term production bottleneck may change from shift to shift
or day to day due to the degradation of various produc tion equipment and dynamic
fluctuation in buffer status. The identification of transient production bottlenecks
answers the question “which machine failure most seriously endangers the production
schedule?” Answering this qu estion in a systematic way facilitates the maintenance
prioritization decisions.
6
As we said there are always conflicts between the production manager and the
maintenance manager. To address this issue of how could one perform systematic
equipment maintenance during production without affecting the production through-
put at the end of the line, it is necessary to accurately estimate the available
opportunity window for maintenance in the midst of production runs. By analyzing
the system configuration, current buffer status, and equipment conditions, it is
possible to identify sufficient machine time for the preventive maintenance without
interrupting normal production schedule.
7
There is a tradeoff between maintenance personnel staffing levels and the
throughput of a production line. The more reactive the maintenance personnel is,
the lower the probability that a given repair will be delayed waiting for available
personnel. On the other hand, increasing the number of personnel increases the labor
costs. Therefore, one important goal of IT-based maintenance systems is to perform
adequate maintenanc e with minimum maintenance crew size. By constructing an
optimization formulation, one can solve this maintenance staffing management
conflict.
8
The algorithms for solving this issue as well as the one that identifies
sufficient machine time for preventive maintenance actions without interrupting the
normal production schedule were implemented at General Motors.
*
*
Three “BOSS” Kettering awards have been given by General Motors for the successful implementation of
the research conducted by Professor Jun Ni and his students Q. Chang and Z. Yang.
IT-BASED MAINTENANCE OF LARGE SYSTEMS 329
The system-level controller is the best one for making effective maintenance
decisions. It is positioned where it can receive information from the machine and the
cell levels that is sent up to it. These inputs can then be compared to the overall
production requirements that are sent down from the enterprise level. Implementing
such IT-based maintenance in large manufacturing systems boosts their productivity
and increases their reliability and responsiveness to changing operations.
PROBLEMS
12.1 Give an example (preferable from your own work experience) on how
supplying the right information to a workforce (at any level) can assist in
responding to unpredictable events and/or to increase productivity.
12.2 Which sensors can be included in the equipment at each level (machine, cell,
system, factory) of the factory hierarchy (Figure 12.3)?
12.3 What kind of specific information is needed to make more effective decisions
at various levels of an organization of a manufacturing company? Give
examples.
12.4 Some executives say that the empowerment of employees contradicts the
traditional hierarchical management structure . Do you agree?
REFERENCES
1. P. Kidd. Agile Manufacturing. Addison-Wesley, 1994, p. 107.
2. R. Slater. Jack Welch and the GE Way. McGraw Hill, New York, 1999.
3. G. Morgan. Creative Organization Theory: A Resourcebook. Chapter 27, Sage Publications,
Newbury Park, 1989, pp. 64–67.
4. Y. Koren. Computer Control of Manufacturing Systems. McGraw Hill, New York, 1983.
5. Z. Yang, Q. Chang, D. Djurdjanovic, J. Ni, and J. Lee. Maintenance priority assignment
using on-line production information. ASME Transaction, Journal of Manufacturing
Science and Engineering, 2007, Vol. 129, No. 2, pp. 435–446.
6. L. Li, Q. Chang, J. Ni, and S. Biller. Real time production improvement through
bottleneck control. International Journal of Production Research, 2008. DOI: 10.1080/
00207540802244240
7. Q. Chang, J. Ni, P. Bandyopadhyay, S. Biller, and G. Xiao. Maintenance opportunity
planning system. Transactions of the ASME, Journal of Manufacturing Science and
Engineering, 2007a, Vol. 129, No. 3, pp. 661–668.
8. Q. Chang, J. Ni, P. Bandyopadhyay, S. Biller, and G. Xiao. “Maintenance staffing
management. Journal of Intelligent Manufacturing, 2007b, Vol. 18, pp. 351–360.
330
IT-BASED ENTERPRISE ORGANIZATIONAL STRUCTURE