Quality-Based
Problem-Solving / Process Improvement
| C Charts | Deming | Juran | VA | Teams | TQM | ISO/QS9000 | DOE | Brecker PI |
|
|
Quality-Based
|
|
Many of today's problem solving and quality improvement tools
Control ChartsStatistical Quality Control (SQC) or Statistical Process Control (SPC) for repetitive, high volume production began in the 1930's when Shewhart developed control charts. Small production samples were measured periodically to monitor quality. Sample mean (Xbar) and range (R) charts were used to detect when a process was going out of "economic control."
The causes of variations that exceed the upper and lower control limits (UCL and LCL respectively), such as at "A", must be eliminated in order to bring the process back into statistical control. Deming PDCA
The Deming / Shewhart Cycle was especially useful in solving these quality
problems.
The PDCA Circle denotes continuous improvement by repeating the
basic
cycle of
In the last 20 years, many companies have adopted similar problem-solving models -- often with 8-10 steps or more. Special attention is given to solving process problems -- problems that occur repetitively. Process improvement models add such steps as assigning process owners, forming teams, and establishing process measures. JuranJuran focused on quality control with the "trilogy"
An SPC representation of the trilogy is used in the figure at the
right.
A
spec
is
developed -- usually with a
When the total quality management movement (TQM) gained prominence
in
the
1980's,
the product planning step was expanded to include
a
broader
evaluation of
customer
needs
(similar to
VA and QFD
).
Value Analysis TeamsIn Value Analysis, developed by Larry Miles at GE during WWII, multi-functional teams (design, production engineering, purchasing, quality) use a formalized process to identify alternative materials, manufacturing processes, and designs to improve function while reducing costs.
The "Job Plan" for a Value Analysis study of a specific product or process has
the following format
The brainstorming methodology in VA has been widely emulated in process improvement teams. Ideas are generated in a non-critical, free-flowing process. Then, ideas are grouped using affinity diagramming techniques and refined as needed for practicality. TeamsJapanese Quality Circles demonstrated the effectiveness of worker teams in identifying and solving process problems in their work area. However, most serious quality problems in non-manufacturing (as well as manufacturing) organizations arise in activities that involve more than one department / function. Quality Circles has evolved into Kaizen, which utilizes multi-functional worker and production engineering teams to improve quality and productivity in a given process. The teams use TQM techniques in implementing "Lean" manufacturing methods. TQM
Total Quality Management (TQM) emphasized using
multi-functional
teams
(professional staff and workers from all departments involved) to solve
problems. The teams
were
trained to use basic statistical tools to collect and analyze data.
ISO / QS9000ISO9000 was developed as a standard for business quality systems. To be certified, businesses needed to document their quality system and insure adherence to it with reviews and audits. A key element was the identification of non-conformances and a Corrective Action System to prevent reoccurences. Specific quality improvement methodologies were not prescribed. The automotive industry adopted the QS9000 standard for their suppliers to require the use of specific practices in quality planning and in production operations.
Design of Experiments (DOE)Six Sigma process improvement emphasizes getting quantitative data on the effect of key variables in production, service, or administrative processes. Many process improvement efforts go astray because people assume they know all of the key variables -- key variables and especially interactions between variables are not always obvious. Simple statistical data gathering or testing can be used to verify or determine these key variables.
DOE is especially useful in simplifying the improvement process for complex technical and administrative processes.
Brecker Process Improvement
Brecker Associates integrated the
Phase 2: Process improvement solutions are identified and quantified.Value Analysis techniques are used to collect business, product, and process data on productivity, quality, and costs. Workshop participants include marketing, design, operations, quality, purchasing, service personnel as well as suppliers, workers, and customers as appropriate. Customer requirements are determined and quantified using QFD techniques. Process mapping (from TQM) is used to develop an understanding of the product / service and the processes used to produce and deliver it. Quality issues are raised and root causes are sought. VA brainstorming is used to identify potential improvements. Ideas are refined and the benefits of the most promising improvements are quantified. Specific product and process re-design projects are selected for Six Sigma Leaders and / or teams to undertake in Phase 3. Phase 3: Multi-functional teams improve key products and processes.
Design and operations teams (with other members as appropriate) are charged with re-designing specific components and / or processes. They examine the product and process information in detail. Critical-To-Quality parameters are determined for products and for processes. DOEs may be run to identify CTQ parameters and assign target values. Process capability data is obtained. Additional productivity and quality tools, such as
|
Since 1948, engineers at Carnegie Tech (now Carnegie Mellon University) have
been trained to apply the Carnegie problem solving methodology in design. This
methodology parallels the Deming PDCA cycle with the addition of an initial
step to "Define the Problem."
Flow charts or process maps were used to visualize the flow of product or documents through a series of process steps. The predominant goal of Process Improvement Teams was to eliminate the non-value adding steps and to resolve quality problems in order to reduce the (cycle) time needed to complete the process.
In DOE, a series of experiments (tests) are conducted to determine the relative importance of the key variables and to assist in selecting optimum operating values. The number of variables is usually limited so that the time, effort, and cost of testing is not excessive. After the array of tests (experiments) are complete, graphical techniques -- such as at the right -- can be used to illustrate the results.
