Manufacturing intelligence (MI) is the practice of using technology to systematically capture, store, analyze, and act on all of the information gathered by your system. That’s the idea in a nutshell, but it comprises a broad umbrella of perspectives, techniques, and tools that empower it. This leads to some ambiguity as to what people are referring to exactly when they say something “works with your MI” or “builds MI”.

‍

Let’s break down the most essential aspects of manufacturing intelligence and how your business can implement them.

Data gathering for manufacturing intelligence

The first step required to build manufacturing intelligence in your business is to gather the necessary data. There’s a lot of potential useful information that your shop is continually producing, and you need to harness it.

‍

It can be helpful to think of this data in three primary categories:

Manufacturing process data

This is the data generated as you complete the actual work of manufacturing products. You likely have computer-integrated machinery on your production line. You can have these machines report on their output rates, their current settings, etc. These reports can be logged in a central database.

‍

More advanced machines can use technology like computer vision to gather more information about their process. They can visually assess their outputs and judge their quality, thus generating data around defect rates.

‍

This data can be used to estimate the time and quality of future work, and what machine settings are most applicable to each job. This is an essential part of manufacturing intelligence: enhancing the productivity of your process through strategic analysis.

Engineering data

Perhaps the largest amount of data falls into the category of engineering data. This is everything required to begin production of a product: the design, drawings, material information, process information, estimated costs and timelines.

‍

After the work is done, engineering data also comprises a lot of information you can gather about the work in aggregate: how much in total was produced, what the defect rate was, how long each stage took, and more.

‍

This is data that you likely already produce using computer software and thus is stored digitally somewhere automatically. However, the way this data is stored may not be conducive to manufacturing intelligence. Keep in mind that just because data exists somewhere, doesn’t mean that it’s a useful or actionable asset.

Logistics data

Logistics data comprises everything that connects your overall business operations with the specifics of your manufacturing process. This includes the storage and transportation of your products, the human resources of people who make production happen, the overall financials of your business, and much more.

‍

This is likely information you already have well-organized and managed. It’s data that inherently paints a big picture of the highest level inputs and outputs of your business. However, to create manufacturing intelligence around it, it should be connected to the particulars of every deal and process.

Building data lakes for manufacturing intelligence

In order to craft all of this data into something you can use as manufacturing intelligence, the best way is to gather it all in one place. A system that can consolidate data from many sources and contextualize it with each other is known as a data lake.

‍

The benefit of a data lake is that it can pull in data from many different sources without requiring manual reprocessing or oversight. It’s the only way to keep pace with the amount of data your shop generates every day.

‍

A data lake also allows you to uncover strategic insights across all sorts of categories of data and stages of the production process.

‍

For example, you may have data indicating that certain parts have abnormally high defect rates. But this likely is just a mapping of quality data and specific part IDs. To find the common factor, you’d need to link this logistics data with engineering data, looking at the drawings for these parts and looking for elements they all share. Then you can look at the process data for that element to diagnose exactly what’s going wrong.

‍

Making that connection from the highest to lowest level to answer a specific question can be a lengthy, difficult process involving a lot of manual cross-referencing and searching. A data lake makes it automatic and easy, by linking each piece of data with the central design, and connecting similar designs together. The quality data that instigated the question contains its own answer by linking these other pieces of data, connected using design elements themselves.

‍

Making the most of manufacturing intelligence

Now that you’ve got the data needed for manufacturing intelligence into one place, you can start putting it to strategic use. 

Manufacturing intelligence for strategic initiatives

A good way to approach this is from the top down. Make overarching strategic goals for your business. Here are some examples of the level you can work with:

‍

• Reducing costs for product storage by some percent
• Consolidating suppliers down to some number
• Decreasing the production time of the longest production processes
• Reworking the production processes of the products with the highest defect rate
• Improving the speed and consistency of responding to RFQs

‍

For each of these questions, the answers lie in your manufacturing intelligence. Break down the information you’d need to answer the questions into more and more discrete chunks.

‍

For example, when consolidating suppliers, first you’d need to make a list of all current suppliers. Then you’d want to know which parts you order from each supplier. Then you’d look at the types of parts that each supplier provides. Are there areas where one supplier could handle similar parts currently sourced from another supplier? Who’s offering the best deals for each type of part? Who provides the highest quality for each type of part?

‍

Using your data lake, you can have all of this data laid out for you plainly. Which suppliers are most ideal for each type of part you require will become obvious. This is the true realization of manufacturing intelligence.

‍

As the world’s supply chains and labour markets become increasingly unstable, intelligence is the only way to remain flexible and agile. Rapidly pivoting to plan Bs or even Cs could be necessary on any given day. You need a way to quickly understand and evaluate your options.

Manufacturing intelligence for day-to-day consistency

The other important benefit of manufacturing intelligence is from the bottom up. Manufacturing intelligence means continual monitoring of your shop’s outputs on the most detailed level: every part coming out of every machine.

‍

What this allows is the earliest possible detection system for anomalies and problematic trends. You can be alerted immediately if anything goes wrong. Then, using the manufacturing intelligence setup you’ve built, you can immediately contextualize the issue with your business as a whole.

‍

For example, if a machine’s output is slowing gradually due to wear and tear making it less effective, it may be hard to know when to repair it. Repairing costs money and temporarily reduces output to zero. On the other hand, decreased production slowly eats into your profits.

‍

With manufacturing intelligence, you can follow the ramifications of this slowdown up the chain. You can see exactly how the production slowdown impacts your bottom line and determine when investing in the repair makes most sense.

Manufacturing intelligence with CADDi

CADDi is an AI data platform for manufacturing that serves as a powerful data lake and enables manufacturing intelligence. Our patented similarity search automatically collects drawings with similar design elements. Then it links each drawing with all of the contextual information, from the revisions of its design, to the suppliers of its component parts, to the defect rates when it was put into production.

This gives you the detailed big picture perspective necessary to use manufacturing intelligence strategically. See it all in action by checking out a free demo or walking through our product tour.