Transforming Your Manufacturing Data Into Actionable Insight

Have you ever had too much data?

The manufacturing shop floor is constantly generating data. From production times and quality information to work orders and operation events, every manufacturing floor generates loads of data every minute.

And this is a good thing.

For most manufacturers, the more data you have, the better insights you can glean from it.

However, with so much data from multiple sources, companies can get bogged down sorting through it all, leaving them with more questions and indecision.

But the problem isn’t too much manufacturing data. No, it’s that companies are not consistently utilizing efficient methods to visualize and transform their data into something valuable.

So how do you transform manufacturing data into actionable insight?

DRIP Syndrome: The Dangers of Unmanaged Data

You could have all the data in the world but if it's just a series of numbers on an Excel sheet, you’re probably unable to transform it quickly and effectively. This leads companies to experience DRIP (Data Rich, Information Poor) syndrome where they have the data needed to make smart and informed decisions but they are unable to transform the data into real information and knowledge. If this is the case, the data can’t be utilized to its fullest potential.

Another key contributor to DRIP syndrome is when data sets are separated between systems, departments, and operational counterparts.

Let’s say you have a way to visualize the data from each operation on the manufacturing floor, but the data is entirely siloed. One system handles quality, another handles production, and to top it all off, information is not shared between work cells, making it incredibly hard to cross-validate and verify. In this situation, you lose an accurate view of both the overall performance and a granular perspective of each job.

But it’s not all doom and gloom. There are a few smart methods to optimize how you transform your data into actionable insight.

The 4 Phases of Transforming Manufacturing Data into Information, Knowledge, & Wisdom

Data has an exponential value when processed and transformed. In reality, unprocessed data is just a string of random numbers and is only valuable based on its transformative properties. Knowledge and wisdom are behind each figure, and it is every manufacturer's responsibility to extract and transform the insight they need.

The Data-Information-Knowledge-Wisdom (DIKW) Pyramid illustrates how raw data is contextualized and transformed into powerful and valuable insight.

Let’s explore each phase of the pyramid and see how the transformation of manufacturing data works.

Phase 1: Data

Raw manufacturing data consists of discrete figures in the form of numbers, characters, and values that pertain to events, actions, and histories from the manufacturing operation. Remember, every action on the shop floor generates useful data, you simply need a way to collect and extract it from reliable sources.

But what are reliable sources of manufacturing data?

Ensuring that you have reliable data requires a comprehensive approach. As we saw with DRIP Syndrom above, the data you gather should have several layers of validation and verification built into the process, providing advanced levels of data integrity.

Work instruction software provides manufacturers with several intelligent methods to gather accurate data reliably.

1. Smart Tools & Work Instruction Software

For instance, by coupling work instruction software with a smart torque tool, companies can create an error-proof process that automatically gathers and verifies data from every tightened bolt. Using our ToolConnect IoT platform, every time a bolt is torqued, the right parameters and settings are automatically pushed to the tool while the tool records and stores the applied force. At the same time, the system verifies those numbers with the assembly specifications.

To boost data integrity even further, our work instruction software locks the tool from being used if the user is not on the step requiring the smart tool. This practice ensures that all data is captured at the right moment.

2. System/Machine/Database Integration & Communication

Similarly, companies can also increase data integrity by connecting their various systems together via API (Application Programming Interface). By allowing communication between systems like ERPs, digital work instructions, BI software, and an MES, companies can centralize their work orders, production data, quality numbers, and workforce details. In essence, each system validates each other's numbers, ensuring that all figures are accurate and fit the real-life operation.

Similarly, IoT sensors and devices can be integrated within the connected framework, weaving a complex yet robust digital thread throughout the manufacturing environment. By sharing and gathering data from multiple sources, companies can cross-validate their manufacturing data, ensuring accurate readings in almost all cases.

3. User Input & Smart Forms

While automated tools, machines, and systems provide incredibly accurate data sets, manufacturers should be careful not to forget the human workers on the shop floor. These people are creating value firsthand and can provide a great amount of insight. They also have lots of opportunities to capture valuable data.

However, to maintain data integrity alongside people, it is important to get this data quickly and efficiently. From employee feedback to serial numbers and more, the data needs to be captured in-process.

Integrating digital forms into an operator’s digital work instructions allows workers to capture specific data at the moment it needs to be captured. When the data needs to be entered, the work instruction platform will present a smart form to the user. In most cases, the work instructions will not proceed until the information has been entered. Companies can also place checks and balances that prevent the operator from continuing if the data entered is incorrect or unrealistic.

Similarly, as was the case for Republic Manufacturing, smart forms can be used to create digital checklists that enable inspection teams to capture more quality data while speeding up the process.

Phase 2: Information

Data + Context & Organization = Information

So, you now have reliable manufacturing data. But how do you understand it? At the moment, it’s just numbers. In this phase, we can transform the manufacturing data into readily available information by employing two key practices:

• Contextualization
• Organization

However, contextualizing and organizing data can be difficult if not using the right methods and tools. If the wrong context is applied, or if the data is not organized properly, then data integrity is out the window.

Luckily, we can properly incorporate context and organization through data visualization. By taking the raw numbers and filtering them through a visual interface that organizes the data, we can quickly transform our collected data into easy-to-read and contextualized information.

For example, let’s imagine there are several work orders open on the shop floor. Each work order has a number of operations that need to be accomplished in a specific order so that the next operation can begin. In this case, managers and supervisors need a quick way to monitor this information within their contextualized settings while ensuring it is easy for end users to search and use.

A prime example of how to accomplish this is a KPI dashboard within BI software or Work Order Status Cards within an MES as seen below:

When the user opens the Work Order Report, all production and quality data is automatically sorted for them. They receive a real-time look into each operation, including the status of each job, quantities completed, and units available. This visualization method allows leaders to quickly see the data as contextualized information and understand how products are moving through the production line.

To take this even further and gain more information through data visualization, managers and supervisors can monitor more granular information on an individual operational level.

This includes

• Completed, expected, & available quantities
• Rework and scrap quantities
• First Pass Yield (FPY)
• Operational efficiency
• Current work time, expected work time, & total pause time
• Estimated remaining time

This level of contextualization and automated organization enables businesses to transform their data into easy-to-read and understandable information.

Phase 3: Knowledge

Information + Time = Knowledge

The next step is to use the newly processed information to find out the “why” and “how” behind specific events and phenomena. This is primarily accomplished by combining information with past experiences and events to uncover patterns and trends.

In many ways, the knowledge phase can be described as multiple information sets connected by their relation to one another over time. All manufacturing actions are connected, and time is the main vehicle used to see how each new event or action works within the whole business.

Root cause analysis practices such as the 5 Whys look at information and contributing factors over time to determine the cause of any issue or success within the manufacturing operation.

The theory of constraints, at least in the beginning phases, heavily relies on relational information over time to determine the most limiting constraint and the optimal rate of production. Once the constraint is found, teams can deploy several methods to properly subordinate production and maximize the constraint.

Similarly, no matter the type of standardized processes you have employed, tracking changes, adjustments, and improvements are a major source of information over time that can be converted into knowledge. Every change enacted to a process is an opportunity to gain new knowledge and perspective on what works and what doesn’t.

Phase 4: Wisdom

Data + Information + Knowledge = Wisdom

While data, information, and knowledge all look backward at past events and their relationship to one another, wisdom finally allows us to look forward and make informed decisions about the future. In essence, this final phase of the manufacturing data transformation process is where continuous improvement takes off.

Wisdom is ultimately responsible for the application of all changes and improvements to the manufacturing process. This includes strategic decisions like integrating smarter technologies, empowering the workforce with a worker-centric MES, investing in new machines, improving production lines, and optimizing relations with suppliers to improve efficiency and reduce costs.

However, while the application of wisdom will be different for each business, all wisdom heavily relies on the phases that preceded it. In other words, wisdom always needs to sit on a strong foundation of data, information, and knowledge.

For this reason, it is essential that manufacturers prioritize their foundational data collection methods, ensuring that they can easily contextualize that data over time and turn it into continuous improvement that will have a positive impact on future actions.