Statistics for Processes
Statistics are essential for designing, controlling, and improving processes. Fortunately, using statistics is much easier than most people believe.
Statistics – practically:
Sound expertise in statistics combined with practical experience and objective-oriented pragmatism gets you nearly everywhere.
It’s about bringing transparency into problems and making the unseen visible in your processes. We boost product reliability based on facts, you simply were not aware before.
- Analyze correlations between process settings and yield, leading to reduced scrap rates
- Visualize and monitor the variation of your processes to enable fast validation, simple deviation management and increased credibility during audits
- Continued Process Verification (CPV), Ongoing Process Verification (OPV), Statistical Process Control (SPC)
- Optimum parameter setting e.g. with the use of Design of Experiments (DOE)
- Systematically identify root causes for product failures with Design of Experiments (DOE)
- Understand your product performance and weakness in early development phases
- Setting Tolerances – building on customer needs instead of process capability
We adapt the most helpful statistical tools to meet your needs. And help your organization to achieve maximum benefit from this.
Design of experiments (DOE) is a powerful tool for determining specific factors affecting defect levels in a product. When a medical devices company tested their upgraded surgical system used for life-saving operations, previously unknown system crashes occurred. Together with the client’s team of experts we applied DOE and identified the cause of the failure. Unexpectedly, it turned out that these crashes were due to an updated processing unit of a new batch of mainboards. While on paper the specifications of old and new mainboards were identical, replacing this unit solved the problem and the deadline could be met.
We designed and implemented Continued Process Verification (CPV) at a mid-sized pharma manufacturing company to improve quality KPIs while meeting regulatory requirements. Training and coaching involved all employees applying statistics. The results showed the most efficient ways to save time, led to a deeper understanding of processes, improved cross-functional collaboration in investigating root causes and ultimately led to real business outcomes such as reduced deviations and scrap.
Changing a product comes with risks, and test results are sometimes hard to interpret. This automotive project helped a first tier automotive company to systematically identify and follow-up on potential root causes of a prototype development failure by methodically leading the team and by statistically ‘crunching’ test data.