Impact of digitalisation on Cost of Quality (CoQ) & Cost of non-Quality (ConQ)

As from the second half of the 20th century, the quality function has become increasingly important within organisations, thanks to the work of quality gurus such as J.M. Juran or A.V. Feigenbaum. In the past few years, with the increase in quality standards, stricter customer requirements, and the rise of new technologies to reduce the costs of quality, a new perspective on quality has emerged.

Quality management is more than ever seen as an essential part of manufacturing activities, helping organisations to produce cost-efficiently, but customised for and delivered fast at the inbound gate of their customers.

However, measuring and reducing the cost of quality (CoQ) or non-quality (ConQ) is not an easy challenge. This article examines the various factors influencing the total cost of quality, and how that cost can be reduced thanks to new digital solutions.

In order to quantify the quality costs, it is essential to understand what they are composed of. While some costs of quality or non-quality are straightforward and easily measurable (e.g. scrap, rework or warranty costs), others might be much more difficult to identify (lost sales, high inventory, lost customers, lost opportunity, image degradation, etc.).

There are several possibilities to standardise and classify quality costs. In this article, we will separate them into two main categories, as shown in figure 2 below: the cost of quality and the cost of non-quality.

The cost of quality (CoQ) can be defined as the cost of providing products and services as per the required standards. They comprise the prevention costs, associated with avoiding the occurrence of a defect, and the appraisal costs, spent to detect if the product is manufactured according to specifications.

The cost of non-quality (ConQ), on the other hand, is the failure cost associated with a process not being operated as per the required standards. The failure cost can be either internal, when the failure is detected before products reach the customer, or external, when the failure is detected after a product is shipped to the customer.

Figure 1 - The hidden costs of quality (iceberg model)

Figure 1 - The hidden costs of quality (iceberg model)

External failures almost always incur the highest cost of quality impact since these types of failures might cause bigger trouble, such as an impact on brand image or the recall of a product.

It is to be noted that the total CoQ is usually expressed in relation to the total revenue. “Many organisations will have true quality-related costs as high as 15 to 20 percent of sales revenue, some going as high as 40 percent of total operations. A general rule of thumb is that costs of poor quality in a thriving company will be about 10 to 15 percent of operations” (ASQ, 2019).

Figure 2 - Classification of the types of quality costs

Question

As from all the different types of quality costs that exist, it is clear that measuring the total cost of quality is a real challenge. Therefore, many companies do not include all aspects in their CoQ calculations. Costs left aside are for example the costs of re-inspecting products and dealing with customer complaints or the failure costs (CQI, 2018).

Nonetheless, those companies still face a very high CoQ, compared to the best in class manufacturers. The question that arises is then how it would be possible to lower the CoQ and how digitalisation can be part of the solution.

"PwC didn’t simply sell us a device but designed a solution based precisely on how we operate. We could immediately put the solution to use and see how it worked and the possible savings it could help us make, with no investment from our side, meant there was no risk in trying this new and innovative solution."

Raf Van de Put Facility Engineer, Philips Lighting

Our conviction

We believe that a lot of quality problems, and thus quality related costs, arise because of a lack of information or because the available information within an organisation is not properly used. Indeed, in the digital world we live in, some businesses are overwhelmed by data, and it is not easy to know whether that data is relevant or not, and how it can be analysed in order to associate it with the right performance output.

In other businesses, it might be that the digital transition is still on its way, and that the data is still not collected at its full potential. According to a joint study of ASQ and Forbes, two thirds of the companies lack data for driving and measuring quality initiatives.

Figure 3 - All external and internal data for driving and measuring quality initiatives available today
Source: ‘The Rising Economic Power of Quality - How Quality Ensures Growth and Enhances Profitability’ (ASQ & Forbes, 2017). Note: Does not total 100% due to rounding.

Figure 4 - The impact of an investment in prevention on the total cost of quality based on Juran’s model for optimum quality costs

Figure 4 - The impact of an investment in prevention on the total cost of quality based on Juran’s model for optimum quality costs

In any case, solutions exist to collect, share and use information properly, in order to avoid quality issues. By collecting and processing production data, it is possible to set up a prevention system efficiently and easily.

As shown in figure 4, as a result of this investment in prevention and proactive quality, there will automatically be fewer defects and, accordingly, a reduction in the cost of internal and external failure. Moreover, the gathered information can be used to help control the quality of the products in a more effective way, which will have a positive impact on the appraisal costs.

In order to reach such an efficient quality organisation, new technologies arising in the digital age can help companies to gather data and use it efficiently.

Solution

This part of the article covers in more detail some of the digital solutions that enable CoQ reduction. First, the focus will be on the new technologies enabling companies to gather data efficiently and to share it with the people who need access to it.

The second part explains which new solutions can be used to analyse the available information and draw insightful conclusions, which will allow quality cost reductions to be achieved. Additionally, some practical cases will be presented, as examples of how those solutions can be put into practice in the real world.

 

Solutions to facilitate access to information

Collecting and having access to the relevant information is key for quality professionals. Sensors and the Internet of Things (IoT) can help. When thinking about IoT, many people think about connected products like a smart watch, a connected fridge that tells you what to cook or what to buy in the grocery store, or a coffee machine that starts making coffee when your alarm is going off. In fact, IoT is a lot more than that. Indeed, another aspect of IoT, outside the consumer domain, is the Industrial Internet of Things (IIoT).

The IIoT links machines, sensors, computers and software to the internet and to each other, enabling the collection of data to improve business performance. Thanks to the network of digital connectivity, industrials can collect production data that can then be analysed and used to improve the processes and predict and prevent problems. How digital solutions can make the data analysis easier and more efficient is explained further in this article.

Another application of IIoT is to enable industrials to track business performance and display it in the form of dashboards or teamboards, enabling visual management of the information. Sharing insights about when or why errors happen with the operators will prevent those errors from happening in the future and will install a culture of continuous improvement at the shop floor level.

Additionally, IIoT makes it possible to monitor the production lines in real time and can be used as poka-yoke systems within the process. These are mechanisms that help avoid mistakes in a manufacturing process. The connected sensors can detect any defects and prevent the product to move to the next step in the process if it is not ready or if it is deficient. For example, if there is a need to check the presence of the required pieces before going into the robotic welding process, proximity sensors can detect the presence of the different pieces and stop the process if one or more are missing.

Industrials investing in IIoT will thus, among others, be able to collect production data, to track and display their business performance in order to take corrective and more efficient preventive actions and to introduce poka-yokes inside their processes. This will enable them to set up more efficient prevention systems and to avoid more failures, consequently lowering both the internal and external failure costs. A more efficient and automated prevention also means IIoT enables prevention cost savings in the longer term. Through 2020, industrial companies are forecast to invest $907 billion each year to their IIoT initiatives. They expect that these investments will lead to $493 billion in increased revenue and to $421 billion in reduced costs between 2016 and 2021.

Figure 5 - Spending on Internet of Things worldwide in 2015 and 2020

Figure 5 - Spending on Internet of Things worldwide in 2015 and 2020
Source: ‘2017 Roundup Of Internet Of Things Forecasts’ (Columbus, L., Forbes)

Solutions to use information efficiently

Even if quality professionals succeed in gathering the right data, they need to go a step further and to be able to learn from that data. Generating interesting insights that will enable decision-making is not easy. Fortunately, new technologies exist to help with that. Artificial intelligence (AI) enables computers and other devices to learn from their environment. It allows machines to sense the environment, to think, to learn and to respond and so to contribute more intelligently to business activities in an autonomous way.

Consequently, AI can reduce the amount of routine work humans need to do. AI is in fact a generic term for all “smart” technologies that are aware of and can learn from their environment. Those technologies include machine learning and robotic process automation (RPA), which are interesting in the context of reducing the costs of quality.

Machine learning allows computers to learn from data. This means that, thanks to this technology, computers are able to recognise patterns and predict outcomes, generating interesting insights from the data that has been collected. This enables industrials and quality professionals to conduct thorough root cause analyses (RCA) by linking the data and the quality outcomes, so that not only the symptoms of the quality issues but also their root causes are identified. When correlations are found between failure occurrence and independent variables, the computer will learn to predict when a failure will happen in order to be able to avoid it. It can even suggest (prescriptive) actions without being explicitly programmed.

In essence, thanks to machine learning, the data analysis can be performed fully independently, with less risk of error and in real time by computers, while it would have been very time-consuming and less performant had it been done by humans with more traditional techniques. This makes the prevention system not only better but also cheaper on a daily basis in the long term, as it frees up time for quality professionals to do other tasks, and decisions can be taken faster. Moreover, with a better prevention system that is able to learn by itself, there will be less and less defects, and the failure costs will automatically decrease.

For instance, PwC helped a client in the process industry to set up automated disturbance detection using machine learning. Cameras are recording images of the production process, and the captured data enables automatic, vision-based anomaly detection through a deep-learning neural network. The detection system warns the operators when a failure occurs (even in advance), realising a decrease in down time and waste, as the operators are more proactive and responsive.

When used in combination with robotics, machine learning can also result in a very effective and cost-efficient appraisal system. This is because controlling the quality of goods can be fully automated, with damaged products being discarded on the basis of a computer decision. Investing in such an autonomous appraisal system will reduce the likelihood of errors (that inevitably happen when humans perform the task) and therefore lower the external failure costs. Furthermore, having this control process automated means that less resources are needed for this specific task, lowering the appraisal costs.

However, it is not easy to completely remove human intervention from the appraisal system. Besides, full automation is not always currently possible, and it will not always make the appraisal activities more efficient or cheaper. This is why, in some cases, it is more interesting to combine human power and computer power, for instance with intelligent wearables. This has been proven to be a good solution for Philips Lighting, which PwC supported by proposing a tailor-made solution in the form of a mobile app.

Philips lightning machine manufactur

The app allows Philips Lighting’s utility operators to monitor equipment and to carry out required preventative maintenance tasks and inspection activities. Philips Lighting has calculated that it has made a noteworthy six percent saving in total utilities maintenance costs at the facility as a result of the wearables solution.

RPA, as other interesting AI technology, can automate repetitive business or industrial tasks. This technology can be very helpful for all administrative work related to quality. This is because, thanks to RPA, it is possible to set up many different kinds of automated systems, such as claims management, reporting or classification. All those automated systems can help to lower the external failure costs but also all costs of quality in general.

Besides, they enable a company to get rid of most paperwork, which also represents some important savings. For instance, PwC supported a steel processing company to move to paperless quality management, as the company was making intensive use of paper, making track & trace of issues and statistical process control (SPC) almost impossible. As a result of that support, the company knew exactly which tools to use for achieving a paperless and efficient quality process

Conclusion

The use of those new digital solutions in industry, also referred to as “Industry 4.0”, will undoubtedly have an impact on the quality function within an organisation and the related costs. As the companies are shifting their costs to prevention when investing in new technologies, the total CoQ becomes lower, because fewer defects and failures mean decreasing appraisal and failure costs.

Besides this, the speed of Industry 4.0 and the high quantity of available data enable organisations to accelerate their responses to issues or changes, with an increased emphasis on customer experience.

Ultimately, all critical quality points could be monitored with inline sensors to track and manage non-conforming products and to automatically alter machine and resource processes in real time, as analytics are used to predict quality issues before they occur and to take corrective action and to update systems and databases in real time.

However, arriving there takes time, and technologies such as IoT and AI still raise some concerns, namely about security, data privacy, computer infrastructure and job creation/elimination. Especially when it comes to quality processes, there is a high risk aversion regarding the adoption of new technologies. Yet, PwC is convinced that IoT, AI, robotics and other technologies will reshape the world in the coming years, as they are part of the “Essential Eight” core technologies that matter most for business, across every industry.

In essence, companies need to remember that, regardless of the current level of digitalisation of their quality function, it is of prime importance to create a real quality culture inside the organisation. This means that quality will be more embedded in operations, and that there will be a closer collaboration between IT professionals and quality professionals.

 

Authors of this article

Yigit Ciniviz, Senior Associate

Yigit Ciniviz

Senior Associate

Simon Matéo, Senior Associate

Simon Matéo

Senior Associate

Alice De Wilde, Associate

Alice De Wilde

Associate

Sources
ASQ. (2019). Cost of Quality (COQ). Retrieved from https://asq.org/quality-resources/ cost-of-quality.
Forbes. Columbus, L. (Dec. 2017). 2017 Roundup Of Internet Of Things Forecasts. Retrieved from https://www.forbes.com/sites/louiscolumbus/2017/12/10/2017-roundup- of-internet-of-things-forecasts/#2d3aa4fe1480.
CQI. (2018). Quantifying quality costs in the digital age. Retrieved from https://www. quality.org/article/quantifying-quality-costs-digital-age.
Forbes Insights. (2017). The Rising Economic Power of Quality - How Quality Ensures Growth and Enhances Profitability.
Juran, J.M., De Feo, J.A. (2010). Juran’s quality handbook - the complete guide to performance excellence. 6th edition.

Contact us

Peter Vermeire

Partner, PwC Belgium

Tel: +32 9268 8064

Jochen Vincke

Director, PwC Belgium

Tel: +32 9268 8055

Follow PwC Belgium