Supply Chain Optimization
In a Project Management Focused
Organization
By
Steve Heidtke
Manufacturing by Design
Industrial and Manufacturing Engineers have
specialized training for improving processes on the shop floor which eliminate
waste and reduce cycle time. While commonly viewed as narrow in scope, these
same techniques can be applied to all functions of the organization. These
individuals already employ the skill set and experience needed to carry out
improvements in Business Process Reengineering, with an emphasis on improving
the supply chain.
My proposal is to employ IE’s to lead
efforts in applying the principles of Reengineering to certain functional areas
that will become working examples to other entities. These “pilots” are where
people can learn first hand how to apply these techniques in their own areas,
whether internal or external, product or service. I begin by reviewing a few of
the obstacles and history of how current techniques came to be. I also present
an overview of some of the techniques employed which may have familiar
terminology. I will conclude by presenting how these techniques can be applied
to the present situation and have potential to improve project delivery.
Contents:
Project Management Organization
Cycle Time vs. Lead Time
Enterprise Resource Planning and Reengineering
Smoothing Demand of Resources
Activity Based Costing
Concurrent Engineering
Process Mapping
Theory Of Constraints
Measures (Metrics)
Internal vs. External Efforts
Pilot Program
Management Commitment
Bibliography
Project Management Organization
Many orders for custom mechanical
systems are accepted with challenging delivery schedules. The schedule can be
broken down into these basic milestones: definition of concept, assignment of
project personnel, development of specifications, review and approval of
specifications, detailed design of system, procurement of components, assembly
and test, customer acceptance, and installation. This proposal should enhance
current project management techniques rather than replace or change them.
Operations in a project management focused
company are of a typical functional organization, with a vertical reporting
structure. Project managers are challenged with delegating tasks to functional
personnel who do not report to them. Department managers and supervisors are
tasked with pursuing the objectives of multiple high risk projects with limited
resources and difficult and conflicting deadlines. This is complicated by the
lack of correlation between project success and compensation to operations
personnel. Much is accomplished through pressure and because it is “the right
thing to do.” Often the project manager needs to go to higher levels in order
to get priority scheduling, which creates conflict between departments and
projects.
Lead Time vs. Cycle Time
Lead time and cycle time are terms
that are often used synonymously. There is a difference, however. Lead time is
a scheduling tool, which is arbitrarily set by a planner and used to determine
when resources need to be allocated to a product in order to finish it by a
specific date. Cycle time, on the other hand, is a measure of productivity
improvement. Lead time is a fixed value which includes all activities in a
process whether they add value or not, such as waiting (queue), transport, set
up, inspection. Cycle time is a moving measure of value added time as a
percentage of elapsed time. Measures of cycle time often reveal 0.1% to 5% of
value added time in the process. 2-5% is actually considered good in most
cases. Imagine the opportunities that would present themselves if only one
quarter of this waste were removed from the process. Lead time reduction
efforts often result in expediting and allocation of resources from regularly
scheduled activities. This begins a cycle of working only on what is late or “hot”.
Cycle time reduction is relatively easy to measure and results in systemic
improvements that allow on going lead time reductions. Since managers tend to
measure improvement in terms of dollar value, it is important to show value in
cycle time reductions. Here are a few
factors to consider:
Inventory holding cost
Idle resources with fixed
cost
Customer satisfaction
quantifiable in accounts receivable
Penalty fees
Lost sales
Floor space
Working capital
Here is an example of how cycle time
can positively affect working capital. If materials on a $10,000,000 machine
were supplied at a steady rate (for ease of example), then the average
inventory level for that job is $5,000,000 for the length of the project. Since
a typical project of this magnitude takes 12 months, by reducing the cycle time
by merely 20%, $1,000,000 less in working capital is required on this project
alone. It can hardly be argued that reduced cycle time does not result in cost
savings and potentially increased revenue and margins for all entities
involved. Perceived productivity improvements that increase cycle time must be
avoided. Therefore, cycle time must become the most important measure in
process improvement.
Enterprise Resource Planning and Reengineering
The ERP system is the information
backbone of the business enterprise. It should be thought of as an
informational hub to other systems. ERP software should follow a generic model
that allows integration with multiple third party software modules.
Reengineering should not be perceived as an effort to make processes fit the
ERP system.
Smoothing Demand of Resources
The current situation involves many
projects in contention for the same resources, particularly at the end of
accounting periods. This is the common push to ship projects by the end of a
quarter in order to realize revenue to meet prior established targets. Once
this cycle has been established it is difficult to get out of. I propose that
reducing product cycle time can reduce the need to push jobs through, by
creating a system that has higher velocity and is more predictable.
Activity Based Costing
Because historically a large portion
of product cost was incurred in manufacturing by direct labor and materials,
much effort has been allocated to reduce these costs. In the past, the amount
of direct labor and material cost correlated closely with the level of support
functions used. Because of this, labor and material cost are used as cost
drivers for allocation of overhead. This system worked well for many years as
long as this correlation existed. This focus has driven most labor out of
product cost, while the cost of “overhead” in the form of indirect functions
and capital depreciation has increased. Ultimately the correlation between
labor and the amount of overhead has been reduced, while the cost model
continues to be used. This causes business decisions to be made on an obsolete
accounting method. For example, a new machine is purchased to reduce direct
labor input to a family of products such as hydraulic actuator end caps. The
machine costs over $1,000,000, and is justified on labor savings measured at a
burden rate of $70 an hour. In order to obtain mandated payback and return on
investment, labor must be reduced dramatically. Once in place however, the
machine depreciation is charged to the entire machining department, which
drives up the burden rate across all product lines. Since the product produced
on the machine now has very little labor input, not much of the depreciation
gets allocated to those products, while other products run on manual machines
carry more of the burden. This reinforces a situation where make-or-buy
decisions are based on erroneous data. Fewer and fewer products need to absorb
an ever increasing amount of overhead in the form of new equipment and business
systems. Efforts to increase overhead absorption by increasing direct /
indirect labor ratio and reducing support functions are an example of how the
accounting system leads to reactive measures. Common practice is to reduce
headcount when short term financial goals cannot be met. This creates a spiral
of a shrinking pool of workers to absorb the ever increasing level of overhead.
Concurrent Engineering
This is one of the most misused terms
in high technology companies. If you aren’t using it, you are a dinosaur. Therefore
everybody says they are using concurrent engineering. A few traits are:
co-located cross-functional groups, design review and approval by manufacturing
engineers, on going training in design for manufacturability, and getting long
lead material on order before designs are detailed. These are key components of
this technique. Using these techniques can have a dramatic impact on lead time
and cost of high value components, and a huge cumulative effect on costs of
smaller custom pieces.
Process Mapping
This is perhaps the most important
step in a process improvement model. Process mapping is extremely important for
understanding the components of and barriers to cycle time improvement. Process
maps are diagrams representing the physical flow of materials and information
through the supply chain. Maps provide an enlightening view of how inefficient
the flow actually is. As an example please refer to the attached process flow
diagram of a single part number through the manufacturing sequence at BF
Goodrich Aerospace (appendix). The manager of the fabrication area was shocked
into action when he saw the flow diagram.
Ironically, it was the push for
efficiency that drove processes to their present day state. During the era of
direct labor reductions and machine efficiency, it was thought that machines
should run non-stop to minimize labor and pay for themselves in the shortest
time. Inventory, queue time, and indirect labor increases were acceptable
tradeoffs in the all important pursuit of productivity. As the cost of space
soared and transportation unit cost declined, machines, offices, and entire
departments have been “shoehorned” into whatever space was available. The
problem is compounded when more space is needed for aisles to move material and
storage of high levels of work-in-process. Office space proliferates as well
due to a lack of understanding that even office functions need effective flow.
With today’s focus on shortened cycle times and reduced inventory, this
thinking no longer holds true.
The specific technique used for
process mapping is not critical. The collection of appropriate data and ability
to put it into graphical form is important, both for analysis and for
rationalizing the need for change to others. Process maps should be made in
each target area for both physical movement of materials (whether product or
service related), and information. These maps become the basis for process
improvement efforts. These maps serve to drive managers into approval of
changes because they visually show the vast opportunities for improvement.
Collection of pertinent baseline data
is important during the mapping, because it gives the ability to set targets
and measure improvements. In a fabrication process baseline data includes: work
in process (WIP), floor space, labor input, machine times, set up times,
product in queue, inspection, and cycle time. In administrative processes,
similar but appropriate measures are used.
Theory Of Constraints
Theory Of Constraints (TOC) is a
double-edged sword. It is useful in simple process models for identifying
bottlenecks. In complex systems it may serve to downplay the impact many small
improvements can have. TOC can be used to identify areas for improvement that
will give immediate and on going value to business processes. An example would
be to measure the amount of WIP or floor space by department for targeting
areas where the greatest gains can be made. The added value (by reducing inventory
or floor space), can be applied to cover initial resources allocated, and used
to pay for additional improvement initiatives. As large initial constraints are
removed, smaller constraints become the target. Constraints which appear
smaller than the resources needed to remove them should be diligently pursued,
because of the lasting and overall effects which are not apparent. If we are to
reach optimum system performance, many small improvements can have a profound
effect. Many of the resources can be thought of as fixed costs (especially
people) and therefore as long as they are allocated to the largest existing
constraint they are properly utilized. One practice to be avoided is lay off of
personnel whose jobs are eliminated through the restructuring involved. Rather,
remove top performers from departments, utilizing them to carry out what they
have learned to the next area for improvement. By removing top performers, two
things are accomplished: average performers are more inclined to assist in the process;
and the better performers are rewarded by becoming facilitators. If management
is tempted to eliminate jobs for short term gain, they will undermine the
extent of improvements possible and encounter resistance as the project is seen
as a headcount reducing activity.
Measures (Metrics)
Selecting the proper measures is
imperative to justifying changes, making systemic improvements, and avoiding
counterproductive activities. For example, measuring machine utilization may
result in larger lot sizes, increasing inventory and creating longer cycle
times. Important measures in process related areas include: WIP (units and
dollar value), cycle time, set up time, overdue orders, floor space, people
(direct and indirect), and average daily demand. Administrative areas have
similar measures, whereas the WIP may be documents or data to be keypunched
rather than units of production. By establishing this baseline data and posting
measures of improvement, focus is placed on these, and not on undermining
effort to show improvement because of concern for job security. This also
serves to show it is an ongoing process and not a management fad where things
can return to “business as usual” after changes are made.
Internal vs. External Efforts
It is widely accepted that internal
efforts are exercised prior to incorporating external entities such as
suppliers in the reengineering process. This serves to give needed experience
and precedence to those who are expected to teach others that this is the right
thing to do. By going through the motions and seeing real results, delegates
from your organization can serve as role models for the improvement process. As
an example, you should establish links between design and manufacturing to
allow for free data exchange, a working model that can be used with outside
suppliers. In one company recently visited (a leader in technology, software,
and engineering resources), designs are created on CAD, printed on paper, then
recreated on CAM to facilitate manufacturing. This activity is wasteful, and
perpetuates poor relationships between functional areas. Procedures need to be
developed and a model created for engineers to follow when outputting data for
use by suppliers. This effort alone could save a significant amount of cycle
time and costs incurred with every project. This is a prime example where the
initial effort results in continuing commercial value.
Pilot Program
Use of a pilot program does many
things: it provides validity to a process reengineering effort, gives an
immediate payback which can be applied to ongoing efforts, serves as a training
ground for personnel to be used in other areas, and uncovers hidden barriers to
change with little risk. A pilot should be selected that has a high chance of
success in a short time. Studies show that from announcement to successful
implementation the project should take no more than 120 days. This is easily
attainable in many functional areas. The pilot should utilize personnel who
presently work in the area and others from a variety of functions, including
the internal suppliers and customers of the process. These others should also
be selected for potential of future involvement in other areas: maintenance,
human resources, accounts payable, purchasing, design engineering. These people
should be the doers in their respective areas, not the managers or lowest
levels. They will become the facilitators in future projects. Involvement can
be a tool to break down communication barriers between functional areas. Having
involvement from a variety of areas creates a culture where we work together to
bring improvements to other areas without concern for “what’s in it for me.” It
takes a team effort to make the type of changes necessary and creativity from a
lot of people, not just those who see the process from inside. As each
following process improvement effort is unleashed, a mixture of experienced and
new facilitators should be involved, until the process has touched every
department in the company. Soon improvements will be under way in all areas,
with clear objectives and an attitude of ability to make things happen.
Following is a list of activities that support a process improvement effort,
which can be applied to most any function:
1. Define implementation team
management
planning
engineering
supervision
technicians
customers (internal)
suppliers (internal or external)
2. Build commitment of team
members
provide a vision for the ideal situation
educate and communicate
define pilot project
demonstrate
management commitment to change (ongoing)
clearly
define expectations
reinforce
competence
3. Establish objectives
target
product / process group
define
scope; how far on supply, demand sides do we implement
meet
customer demand
minimize
WIP (whether product or paperwork)
minimize
cycle time
build
quality into process
implement
pilot within 120 days
4. Define and delegate
activities in support of objectives
process operation chart (matrix showing product, demand,
resources)
collect
baseline data (layout, space, WIP, times, yield, lates)
level
demand (supplier and customer involvement critical)
define
inspection required (make quality part of the process)
reduce
change over times (define, simplify, make external)
define
new layout (simulate on paper, plan future needs)
simplify
planning and reporting (use exception reporting)
define
new roles and train personnel, planners, suppliers, customers
6. Implement changes in a “Kaizen”
atmosphere
coordinate
facilities
move
equipment, offices
“Ready,
Shoot, Aim”
7. Implement visual
management and control
8. Measure performance
against baseline, and ongoing improvement
9. Define, review, and refine
process.
Management Commitment
This is an essential ingredient of
any improvement project. Without higher levels (along the chain of command)
giving the go ahead, people will naturally be reluctant to stay the course.
There is a commitment necessary that people will be allowed the time to carry
out their new assignments, without constant harassment to get their regular job
done. A team participant who is allowed to work on this project in their “spare
time” will always defer to getting their normal tasks done and undermine the
success of the project. Persons who are assigned to process improvement
projects should have their other duties reassigned so this does not become a
deterring factor. The commitment of management must also be on- going and not
only as long as it doesn’t affect short-term needs. The current business
climate of excess capacity is the perfect opportunity to engage your internal
resources as well as suppliers. The more suppliers are involved the better and
willing partners they will be when business picks up. Once process improvements
are in place, future orders will be fulfilled faster, at lower cost, and with
better results.
References
Patrick Northey and Nigel Southway, Cycle Time Management:
The Fast Track to Time-Based Productivity Improvement (Productivity Press,
1995)
Charles Poirier and Stephen Reiter, Supply Chain
Optimization: Building the Strongest Total Business Network
(Berrett-Koehler Publisher, Inc., 1999)
Charles Corbett, Partnerships to Improve Supply Chains (Sloan
Management Review, Summer, 1999)
Supply Chain Management and Planning (Modern Materials Handling, August,
2001)
Paul Thorpe, Concurrent Engineering: The Key to Success in
Today’s Competitive Environment (IIE Solutions, October, 1995)
John Layden, Supply Chain Management Creates Now Roles for IE’s (IIE Solutions, July, 1996)