Manage Risks of Quality-related Data

Data regarding product and system quality, both in terms of quality assurance and quality control, increasingly finds its way into risk management and legal arenas. While the requirements and obligations surrounding quality checks in the glazing market vary by segment, quality data—data related to product and system quality—is an essential business tool and risk management device. Development of a plan to record, retain and evaluate that data can prove essential to quality fabrication.

Understanding the risk from quality data starts with recognizing the difference between well-worn terms like quality control and quality assurance. Quality assurance is a system to ensure correct actions and procedures, while quality control tests the end result to make sure the system is reaching its intended goal. In other words, QA looks to the process and QC looks to the completed product, with both feeding information to each other.

The information that passes between QA and QC represents the fabrication chain of a design, system or assembly. Data collected along the development path can show how something is conceived, constructed and tested. It is also possible to identify insufficiencies, discrepancies and responses to quality concerns. 

The ability to easily identify discrepancies and responses within broader sets of quality data is a tool that we see regularly. Parties challenging design concepts have taken to contesting QA plans as insufficient to ensure delivery of specified building targets, for example, energy, water or sustainability. Challenges to the performance of a product or component are being based on the ability to identify specific QC manufacturing concerns with increasing precision. And, challenges to site assembly are often founded upon a gap in job-checks, at the QA and QC levels, that fail to document necessary performance points.

How to collect quality data

The glazing marketplace must view quality as a big data issue that can be used to evaluate operations and defend claims. To do so, however, requires a process unto itself. 

1 - Identify data to record

Start with considering the quality points that require recording. Each segment of the glazing market has its own unique quality considerations. Those standards and processes likely have points that can be identified and quantified as quality checks within a broader assurance plan. Evaluate those points and decide whether to record. Do not over do; appropriate processes foster quality production without overburdening.

2 - Determine collection specifics and storage

Once the process points are identified, look to the type of information that can be collected and consider its storage. Quality data as a risk management or production tool realizes its maximum use when there is easy and reliable access. Simplify the data recorded without sacrificing essential information. Use consistent terminology throughout quality data, regardless of the person inputting the information. Avoid “free text” entries. Consider whether the process points can be recorded individually (“single cells”) to allow later discrete analysis. And ensure that the data entry and any changes are logged so that improper creation or modification is avoided.

3 - Review the data

Set up a review process for quality data. These regular, scheduled reviews do not delve into what the data means, but simply look at the data itself. Examination of the data can show outliers—points that do not make sense within the broader process. They can also identify incomplete or inconsistent entries. Regular, small-batch reviews of quality data can help avoid later challenges to entire data sets.

4 - Use the data

Be ready to use the data. Develop the tools to respond to business or legal inquires discretely. Mass volumes of QA/QC can overwhelm and eliminate the value in the development and effort surrounding a quality-data system. Without the ability to parse the data, the situation becomes one where quantity does not equal quality.

5 - Determine data deletion schedule

Consider retention. A quality-data plan should evaluate when destruction of data is appropriate. Electronic data storage may ease the obligation of retention. Development of a regular schedule to summarize, evaluate and then destroy quality data—retaining the summary and evaluation—may suffice for certain operations.