Continuous Improvement: The Lean Six Sigma Left-Right Punch
1. Why Lean Six Sigma?
I guess before we talk of the "why" of Lean Six Sigma, we should
briefly address the "what".
Lean operations consist as much as possible of only value added
activities: "Lean" aims to eliminate all waste in the work
place. The outcome is that your operations become high-speed and
low-cost.
"Six-sigma" focuses on achieving consistent product and service
quality by eliminating sources of variation.
"Lean Six-sigma" is the fusion of these approaches to achieve
high speed, low cost and defect free operations, with products
and services that delight the customer and guarantee
profitability for the business. It is necessary because each by
itself is deficient. For instance Lean does not recognise the
impact of variation nor does it explicitly seek customer input.
Six-sigma incorporates customer viewpoints via the Voice of the
Customer tools but fails to specifically identify waste.
2. Lean Concepts: Pull, WIP and Value-Added
The basis of the lean approach is that we produce only what is
required just when it is needed (leading to the term
"Just-In-Time").Lean accelerates the velocity of a process by
reducing all forms of waste.
Pull:
The idea is that the work done in a process should be driven
entirely by customer demand. Each step in the process commences
only when there is a signal from downstream indicating that an
item has exited the process. Implementing a pull system
eliminates overproduction and reduces the need for inventory.
WIP:
Work in process represents the number of things waiting to be
worked on. The speed of the process is inversely proportional to
the amount of work in process.
Lead time = Work in process/Completion rate
Given a desired lead time and a known completion rate (obtained
from observation), the allowable work in process can be
calculated.
Value Added:
This concept has been discussed in our previous articles. Only
work which adds features for which the external customer is
ready to pay is value added. A second category of work necessary
to satisfy an internal customer or regulatory requirements is
known as organisational value added. All others (rework,
multiple approvals, unnecessary movement etc) are non-value
added.
A measure of the degree of value addition is process cycle
efficiency.
PCE = Value added time/Total lead time
In many instances, PCE especially in service situations is below
10% and represents huge opportunities for improvement.
3. Six Sigma Concepts: Variation and Process Capability
Six-sigma eliminates variation by removing the special and
common causes, thus improving the capability of the process.
A fundamental expression of the relationship between a process
output Y and the process variables Xi is:
Y = f(X1 X2 X3 X4 ... Xn)
In attempting to improve the process, we seek to identify those
process variables with the highest impact on process performance
and work to optimise those. The six-sigma methodology follows a
specific sequence with the acronym DMAIC (define, measure,
analyse, improve and control) and tools exist for carrying out
each step of the sequence.
Variation:
An important tenet of statistical process control is that every
measurable phenomenon is a statistical distribution. This means
that the output of a process will generally vary around its
typical values.
Two types of variation exist. Variation from special causes or
assignable variations are caused by conditions that can be
identified and where necessary, eliminated. Common cause
variations are random and cannot be traced to a specific cause.
A process which only has common cause variations is said to be
under control.
The standard deviation of the parameter of interest, sigma
(symbol σ), is a measure of this variation.
Process capability:
This parameter measures the degree to which a business process,
with its current level of variation, is able to satisfy the
requirements of the customer. It is defined as the smaller
difference between the upper or lower customer-tolerable limit
(CL or CU) and the process average (X), divided by the process
standard deviation.
Cpk = min[|X-CL|,|CU-X|]/σ
Acceptable values for this ratio are in the range 3 and over. A
low value indicates that the process is incapable of meeting
customer requirements at a Six-sigma level. The process outputs
will contain defects in excess of 3.4 per million opportunities.
4. Implementing Lean Six Sigma
Many organisations have transformed themselves significantly
through the implementation of Lean Six-sigma. Yet others have
faltered in their attempts at deployment. This is because it is
as much about people and organisational culture, as it is about
specific tools and techniques. To emphasise tools and techniques
to the exclusion of cultural factors is to guarantee failure.
Preparation:
Michael George advises the following sequence to prepare for a
successful Lean Six-sigma implementation.
* Select a Lean Six-sigma champion with leadership and
communication skills
* Establish a baseline snapshot of the organisation with
respect to the business status, existing knowledge and attitudes
towards the Lean Six-sigma initiative
* Interview top management to identify critical elements of
success for the business and for Lean Six-sigma
* Engage key influencers (formal and informal) through focus
groups and interviews, to apply tipping point effects in
rallying support
* Assess the impact of what you learn
Deployment:
In summary, deployment should be carried out as follows using
the DMAIC model:
* Stretch the process by applying lean techniques * Solve the
problem of deviation from standards * Ensure maintenance of the
improved processes using six-sigma
Define: Agree on the problem and stakeholders, ascertain linkage
to corporate strategy and impact on ROIC, agree process
boundaries and metrics
Measure: Establish process baselines, observe the process and
collect data
Analyse: Apply tools like value stream mapping, time trap
analysis, failure modes and effects analysis, ANOVA etc
Improve: Use lean tools like 5S, TPM, Kanban, Kaizen, Poka yoke,
flow improvement, and six sigma tools like hypothesis testing
etc to improve the process
Control: Implement visual management, control charts, process
control plans and the Plan-Do-Check-Act cycle
Our next article considers the application of Lean Six Sigma to
services and transactions.