VALUE PROPOSITION OF BIG DATA FROM A RISK PERSPECTIVE
“It seems like a usual morning in the production plant. Machines are working for 24 hours per day without interruption. Machine workers, engineers and managers have just gathered around the dashboard to talk about production amounts, fulfillment of customer demand and problems of last 24 hours. Although it looks like an ordinary day, today will be more difficult than many others are, as there is an unexpected finding; a spec, which is not measured 100%, but instead controlled with SPC, detected as out of tolerance after 100% measurements of some batches, while investigating another problem. This means; the risk related to sampling has been happened.
Containment actions are immediately implemented to assure that customer receives no defective product. In addition, an investigation has been started to find out and then kill the root cause…”
Does this story sound familiar? It seems like we will hear such stories more often, as the number of important metrics to be controlled in production is continuously increasing with the increasing complexity of products as well as increasing customer expectations.
Statistical Process Control:
Dr. Walter Shewhart developed Statistical Process Control (SPC) during the 1920s. Dr. W. Edwards Deming promoted SPC during World War II and after.
SPC says that there are two types of variation:
- Common cause variation, which is inherent to the process and will always be present
- Special cause variation, which is induced by an external source
Dr. Deming have shown that more than 85% of problems are the result of “common cause” variation. He hold the management accountant for the system and claimed that it is manager’s responsibility to work on reducing this type of variation. Later research puts the estimate at over 94%. Preventing and reducing “special cause” variation is the pre-condition for reaching world-class quality and requires problem solving skills.
SPC Control checks, how stable and consistent the process is. It can be either in or out of control.
- “In-control” - stable and only experiencing systematic or “common cause” variation.
- “Not in-control” – Process is not stable. Mean and variation are changing due to identifiable or “special” causes (usually controllable by those running the operation).
Control charts are based on sampling. Sampling is subject to two kinds of error:
- Type I error (α): “False Alarm” – The sample indicates the process is “out-of-control” but is not
- Type II error (β): “Failure to detect” – The sample indicates the process is stable, but it really is “out-of-control”
Type 2 error is the consumer risk of getting products out of requirements. Moreover, it is the risk of producer of deviating from customer expectation, which would lead to loosing competitiveness and even to serious compliance issues due to legal requirements.
“… essentially, all models are wrong, but some are useful.”
George E. Box (1919-2013), Professor Emeritus of Statistics at the University of Wisconsin
Quality from risk perspective:
“ISO 9000:2005—“Fundamentals and vocabulary for quality management systems” defines quality as the “degree to which a set of inherent characteristics fulfills requirements.”
BusinessDictionary.com describes the quality as “In manufacturing, a measure of excellence or a state of being free from defects, deficiencies, and significant variations, brought about by the strict and consistent commitment to certain standards that achieve uniformity of a product in order to satisfy specific customer or user requirements.”
ISO 9000:2015 Quality management systems — Fundamentals and vocabulary defines risk as the effect of uncertainty. ISO 9001:2015 further explains the concept of risk-based thinking. Risk-based thinking ensures that risks are identified, considered and controlled throughout the design and use of the quality management system of an organization in order to determine the factors that could cause its processes and its quality management system to deviate from the planned results.
Risk management is the process by which we try to control Murphies from crossing a street to starting a business. Why do we bother looking both ways when we cross a street? We look for possible sources of risk like a bus hitting us as we cross the street. The purpose of risk management is the creation and protection of value.
If we switch to a risk perspective, the common definitions of quality become; risk of defects, risk of customer dissatisfaction, risk of uncontrolled process variance, risk of product unreliability, risk of security breach, risk of lack of fitness. In other words, failure to achieve objectives.
There is always a cost related to quality control. This cost should be evaluated considering the risk of a possible damage to reputation, legal liability and potential loss of business as a result of the customer receiving defective goods. Quality 4.0 initiatives might help producers add intelligence to monitoring and managing operations. I believe that big data will give companies great opportunities to reduce those risks at acceptable costs.
Published at pmmagazine.net with the consent of Rüya Demirtaş