Chapter 5: Advanced Quality Management for Dummies
The key to understanding big data’s potential is realizing it is only as valuable as the decisions made using it. Therefore, by monitoring data and understanding its patterns and trends, you can turn data into information and information into intelligence. It becomes the secret sauce that organizations use to outperform the competition and satisfy their customers.
Advanced analytics derived from quality system data gives leaders the information they need to make timely, proactive decisions crucial to their business performance.
This chapter looks at how robust quality data can feed the analytics used to identify trends, make better-informed predictions, and realize continuous improvement.
Examining QMS Analytics
The ability to use advanced analytics has made a significant impact on businesses. Consider the problems that manufacturers face without a good analytics solution:
- Siloed data with no clear indication of what’s important
- Higher costs and delayed revenue from being slow to market
- Inability to discover trends and insights needed for decision-making
- Slow identification of issues
- Inability to make data-driven decisions and take action
- Difficulty in determining root causes, resulting in recurring issues
- Lack of knowledge about best practices
How many of these do you recognize? Solving these problems is crucial to have the information you need to satisfy your customers and run a cost-effective business.
Reviewing data challenges for manufacturers
Advanced data analytics can sometimes make the difference between a satisfied customer or a product recall. To determine if you’re confident that you’re making the best use of your data, ask yourself the following:
- Are you still relying on manual data collection? The volume of data now available to manufacturers makes it virtually impossible to manage and analyze it manually. Without some formal data collection automation tool like a QMS, essential insights can be lost, and you cede the edge to your competitors.
- Are you concerned about whether you can rely on your data’s accuracy? Unless you have a trusted automated solution to manage data, you could be using inadequate or inaccurate data that could skew your findings.
- Does an internal stakeholder own the data collection process? Ensure that you have a specific department or project manager overseeing analytics collection to intervene if standards aren’t met.
Identifying the benefits of advanced analytics
Advanced analytics provide the opportunity for you to take a proactive approach to quality. This ensures you can identify quality issues before the product goes through the production process. Quality is transformed with an analytics solution.
With a quality analytics solution in place, you get:
- Quality integrated with your business intelligence strategy
- Better cross-department communication about quality
- Data integrated from multiple systems, including QMS, MES, ERP, and CRM systems
- Data-driven analysis for decision-making based upon data from across the organization
- Faster resolution, leading to reduced incidents
- Support for continuous updates and improvements
The value of advanced analytics stems from the fact that the data from various sources can be displayed in optimized dashboards based on the user’s needs. In addition, because it’s stored in the cloud, everyone can access it as needed from anywhere, provided they’re granted access. You can then set your sights on preventative measures instead of manually wading through data to find something useful
Tackling Data Volume for Quality
To tackle the large volume of data you collect every day, you need a QMS analytics solution that provides a comprehensive view of the data and how you can use it to create actionable insights. To accomplish this goal, you need four foundational QMS elements that are covered in the following sections.
Viewing a quality operational dashboard
With all the volume of data it collects, your QMS must provide software tools to visualize and organize the data to be beneficial to everyone. You need analytics dashboards that are built on quality management industry best practices. For this reason, look for a QMS that provides dashboards that are:
- Pre-built: You want to have several pre-configured dashboards that let you get started right out of the gate.
- Quality-focused: You want to make sure you have dashboards that are explicitly focused on quality throughout the organization.
- Easily configurable: You want dashboards that can be configured to meet the business requirements of your specific quality operation.
Utilizing a quality data lake
You may need a dedicated data lake in which quality data resides. A data lake pools together different data types and structures from a range of sources, including from other enterprise systems. The pooled data is normalized and optimized for analysis. Your data lake should make it easy to incorporate quality data into a corporate-wide business intelligence and analytics strategy.
Leveraging an insights engine
An advanced QMS should offer an “insights engine” constructed to leverage leading analytics technologies to optimize performance while providing best-in-class data analysis, data visualizations, and dashboards.
Deploying a synchronization data layer
A synchronization data layer (SDL) can pull information from the QMS, transform it into workable data, and deposit it into the quality data lake for analysis with no performance degradation to the QMS. This data layer should automatically stay updated with QMS data even as workflows and processes are updated.
As you can see, these are complex foundational elements that must be present in your QMS to ensure that you have the best solution for your business.
“ETQ Insights gets us really excited because it allows us to pull data directly from ETQ without having to do any data-transformation on it. Now with Insights a business user can quickly extract data and start working with it. This is the basis of the next revolution — data democratization; giving access to our end users and letting them define what they need.” – Joel O’Connor, Johnson & Johnson
Looking at the Value of AI and ML in Quality
The advent of artificial intelligence (AI) and machine learning (ML) has given businesses the ability to analyze the vast amount of data in ways that manual intervention never could. For example, a simple customer survey can yield some information, but an analysis of social media data that’s constantly updated paired with internal customer information (CRM) and business intelligence can reveal patterns and trends not detectable manually. The difference in the value of this data is enormous.
How does AI and its subset ML change the game for manufacturers? These apps take the mountain of data that the QMS extracts from every part of the system and analyzes it for insights, trends, and so on. As it does so, it “learns” about the data, which means that no pre-programming is required. For example, humans could never write programs to analyze the data by predicting what would be found ahead of time. Furthermore, because the data is continually updated, the model is constantly being refined.
Concerning quality, ML and AI play an ever-increasing role in areas including:
- Quality control: Equipment can be trained to spot trends and defects, and trace them back to their origin.
- Supply chain integration: Integrating the supply chain is a significant development for manufacturers. It allows for the integration of planning and the ability to spot shortages and defects before they become a major problem. (For more about supplier quality, see Chapter 2.)
- Predictive maintenance: The use of sensors helps manufacturers spot equipment problems before they break down, avoiding costly delays. (Check out Chapter 4 for more about sensors and Industry 4.0.)
One area where machine learning could provide immediate value in quality and safety management is for intelligent trending, or predictive analysis. Leveraging extensive historical data-sets, professionals can move beyond simple extrapolation of historic counts, and, using hundreds of related data points, actually predict what will happen in the future, with a “cone of confidence” (as opposed to the cone of uncertainty) that conveys the margin of error over time.
Read the full article, “Bringing Quality to New Heights with Machine Learning,” by Morgan Palmer, ETQ co-founder and former CTO.
Each week we will share with you an excerpt of a chapter from Advanced Quality Management for Dummies. If you’d prefer to read the book at your own pace download here: