Manufacturing analytics is a new class of software that brings predictive analytics, big data, industrial internet of things, and mobile-first design to manufacturing companies.
It’s purpose-built to deliver great analytics, provide visibility into the plant floor, and help manufacturers win the day, every day.
Mingo’s goal is to provide visibility, improve efficiency, and increase the productivity of manufacturers around the globe.
When we started designing and building Mingo in 2015, we were motivated by three observations we had on the state of the industry:
From our research and experience, we knew there was a Titanic-sized hole in the marketplace for a manufacturing analytics product; one that is affordable, flexible, cloud-based, easy-to-implement, and easy-to-use. But, what exactly is manufacturing analytics?
As you read through this post you may think it is just another BI (business intelligence) tool, MES, or SCADA system replacement. While there is some overlap between the products, manufacturing analytics is in a class by itself.
What makes manufacturing analytics different is that it is purpose-built to handle the time series data manufacturing companies produce every day. Manufacturing analytics is focused on collecting and analyzing data rather than process control. Data from an unlimited number of sources can be collected and correlated together to identify areas for improvement.
In the past, if you wanted to collect information from operators or machines on the shop floor you would invest in HMIs (human-machine interface), SCADA Systems, MES, Microsoft Power BI, business intelligence, data historians, data logger tools, or any of the many other manufacturing software tools out there. All of which are very complex and expensive to set up and maintain. But now with manufacturing analytics, you can buy a single software package to address your data collection and analysis needs.
Built by people who know manufacturing, Mingo provides the 21st century “Smart Factory” experience that manufacturers need to grow in a modern environment. See how it can help you drive revenue.
Let’s talk about the fact that most traditional manufacturing software solutions are not only complex but very expensive. But they sure don’t have to be, especially when manufacturing analytics is breaking the mold.
Many of the on-premise enterprise platform solutions on the market require an additional cost of expensive hardware or software in order to gather data from the plant floor. Building custom data feeds from machines to ERP systems can cost tens of thousands of dollars on top of the software implementation (and require internal resources and/or expensive consultants).
These custom implementations might pull data from the plant floor, but most don’t provide users with anything close to real-time data visualization or role-based dashboards that could be configured easily for different views of your operations, maintenance, and executive managers.
Connecting data feeds to an ERP is possible, but most of the time, the data is not used for real-time purposes. Often the data collected remains stored in historians or on-premise databases, used sparingly for aggregated reporting or auditing.
In order to use some of the more advanced capabilities of these platforms, you need more and more custom data pulls from your plant floor and more inconvenient data entry from your operators. The immense cost of these systems and their approach to analytics provide no ROI, and the systems themselves made little impact on the productivity of the plant floor. A whole level of visibility is lost despite the sizable investment in an ERP.
At their basics, analytics on the plant floor requires three things:
1. Connectivity with your machines.
2. A place to store your data (we store it in the cloud).
3. A set of algorithms and dashboards to visualize and communicate that data.
The value of powerful analytics with the affordability of a purpose-built solution is unfounded. In the case of the enterprise software cost structures, the data isn’t really the thing, it’s the thing they need to do the more expensive things they are really selling you
Manufacturing analytics was developed to solve the problem of overly expensive, outdated traditional software. It is purpose-built to deliver great analytics, visibility into the floor, and help manufacturers win the day. If you need visibility onto the plant floor, you don’t need to implement it from the top down. You can gain that capability per machine, cell, line, and at the plant level, incrementally if you wish, by monitoring a few machines and only one data point. By investing straight away into manufacturing analytics, you will find value immediately by being able to measure and improve performance.
We’ve seen customers gain visibility into availability and performance by measuring only the downtime reasons automatically collected from their machines. That’s a lot of insight without the cost and complexity of a custom implementation.
One customer uses our software on their printing machine at the end of their customer’s packaging line laser printing labels. The value-add of manufacturing analytics to their end customer is that they now have information on throughput about the ENTIRE line.
Most manufacturers are run by smart people that know that data is important to run their business. It’s why they create standards, collect data, and use that data to drive change on their shop floor. This doesn’t always mean that these businesses deploy true analytics for manufacturing or implement best practices for gathering data inside of their organization, but typically means they are very conscious of its importance.
The reason why some are more mature than others in this area is because of their ability to understand what they truly need to learn from analytics to drive cost-change in their business. This stems from a spectrum of awareness around the truth and the story; as it relates to what’s happening on the shop floor. So, what’s the difference? I’m glad you asked.
The truth = what actually happens on the shop floor
The story = what’s supposed to happen and what you think actually occurs
Let us explain. Most of us in manufacturing clearly realize that there is a chasm between what is supposed to happen on the shop floor and what actually happens. In most cases, we hope that the variables are small and that what we think is supposed to happen is generally what takes place. If it doesn’t we’d know, right?
Well, the more and more we learn about manufacturing, the more the industry is showing that this may not be the case. What’s happening between what was supposed to happen and what actually happens may be a bigger deal for a lot of manufacturers than initially thought.
The maturity in analytics that we discussed earlier is directly determined by a manufacturer’s ability to dig into that massive space in between and affect it. Where do you fall? Let’s find out.
Do you have scheduling issues? Do you know how long it takes to make a part? Are you able to properly account for labor hours and production time? A lot of this is ultimately determined by your ability to accurately understand and affect your scrap rates. Scrap rates influence all of these things directly; yet are rarely looked at with a sharp eye for a number of reasons.
Most manufacturers we’ve worked with often set a certain scrap rate (maybe it’s 4%… just as an example). 96 good ones for 4 bad ones. Easy-to-understand numbers that make it simple to plan and build processes around. The problem with standards is that they can typically only be measured in snapshots and determined under optimal conditions (on paper). And as many of you in manufacturing know, things rarely work exactly how they do on paper and aren’t nearly as efficient.
This is exactly the case with scrap rates. People, machines, and processes can create tremendous variability in this whole process. The inability to look at how things are working as they happen means that these snapshots are only historical perspectives of how things actually played out. So at the end of the year, a manufacturer may find out that they were at a 10% scrap rate but have absolutely no idea why.
What’s worse, if the goal was 4% and the end of the year is 10%, you’re still only looking at the average. It’s possible you were at 4% the majority of the year but were at 30% during a certain period; which could be skewing the data. The truth is you will have no idea why.
Imagine this, you think that for every 100 widgets you make that you make 4 bad ones, but in reality, you make 10 (150% more). That means it takes you longer to make the right number of products, make you run production longer, buy more raw materials, painstakingly change the schedule, and add additional labor all because you had no idea how long it took you to make the right number of items, parts, etc. This is directly related to scrap rates.
Extrapolate that over days, months, and even a year! That is a massive cost that many either aren’t looking at, aren’t adjusting, or aren’t able to affect because they don’t know what’s actually happening.
This is why analytics for manufacturing is so important. The truth and the story are both very important, but neither on their own can save you money. What you need is in-between.
The answer is a lot. At its core, the tool is designed to provide performance and quality metrics. Below is a list of some of the more common scenarios Mingo and manufacturing analytics supports.
Out of the box you can track OEE (Operational Equipment Effectiveness) for a machine, work cell, plant, or the company as a whole. The system can collect data from operators, external sensors, or directly from the machines using sensors already built into the equipment.
From this data, you can measure different types of downtime, shortstops, slowdowns, etc., product trends, and even improve OEE quality. OEE, good parts, scrap, and other metrics can be displayed on scoreboards over machines or work cells in the shop so operators and managers how they are performing.
Tracking and reporting scrap is vitally important to all manufacturing companies. Using a manufacturing analytics solution offers you a lot of flexibility and tools for reporting quality.
Operators can record scrap reasons as they are discovered. Data can be collected directly from gauges and other testing tools to automate the collection of scrap data.
Using trends and algorithms in the application, the system can identify when parts start to deviate from the specification and alert the operators or quality teams.
Manufacturing analytics software gives manufacturers the ability to easily import a schedule from an ERP system or Excel and customize it according to the shift or day’s needs, but importing your schedule isn’t the only benefit. No more manual work on the back end, coordinating jobs and communicating it to the floor.
Scheduling in manufacturing analytics software like Mingo helps manufacturers:
Scheduling doesn’t need to be complex and tedious.
Do you ever find yourself wondering how to layout jobs and meet goals based on the entire line, rather than just one machine at a time? That’s exactly what manufacturing analytics software solves. With Mingo, manufacturers can layout and model a production line, getting information on every station or on the line as a whole, in a single view.
You can also integrate line configurations within scheduling to get a clear understanding of your production schedule.
Manufacturing analytics systems are smart they learn from the data they gather. For example, many companies want to identify bottlenecks in their manufacturing processes. There are a lot of ways to do this but most companies will turn to traditional ERP systems to find the bottleneck, this is problematic for a couple of reasons:
1. The ERP system usually represents the best-case scenario and does not include all the alternates.
2. Many times the routing in the ERP is not what happens on the shop floor.
Using a system like Mingo, the system will learn all the routings for a part and identify the bottlenecks in each routing. It can track and calculate average, minimum, and maximum cycle times for a part by machine, cell, or the part itself helping you identify areas for improvement.
Collecting data is a wonderful thing but how do you use and make sense of it? This is where manufacturing analytics excels.
Massive amounts of data can be consolidated and summarized into easy-to-understand metrics, in real-time. The metrics are combined into standard dashboard components to help you understand what is happening in the shop. The dashboards provide access to data by role so users only see what is important to them and don’t get distracted by a lot of noise.
Scoreboards can be displayed in the shop so everyone can see how a cell, machine, or the entire plant is performing.
As manufacturing analytics has matured, algorithms have been developed to automatically find anomalies and bring them to the attention of the correct people with real-time alerts.
The use of visual management resources like dashboards, scoreboards, and mobile apps increases engagement and communication on the floor.
Combined with automatic, real-time reporting and real-time alerts, manufacturers can see exactly what is happening in the plant, at any given time.
One other almost universal truth about the manufacturing software space, beyond the expensive prices, is that many solutions are needlessly complex.
Manufacturing analytics is not. For Mingo in particular, we used design and user experience best practices to make sure no single screen in the application did more than it needed to.
With manufacturing analytics, you have to start somewhere, and traditionally speaking, starting small can guarantee the success of such a project. We knew the majority of our customers would at first use Mingo to track and monitor machines through one data point: availability. Downtime tells you a lot, so we focused on designing the perfect dashboards for that. Once we nailed downtime, we moved onto performance, quality counts, and scrap, keeping close to the principle that when displaying data “less was usually better” and always keeping our end-users in mind, providing simplified and easy-to-understand analytics.
Essentially, the entire product had to be designed to put manufacturing analytics into the hands of the people who needed to use them and be simple to use.
With that in mind, we built Mingo Manufacturing Analytics to satisfy specific requirements:
Following these simple tenants, Mingo is able to provide low-cost, configurable analytics to help manufacturers improve performance and provide visibility onto the plant floor.
When set up to deliver information like scrap rates in real-time, analytics for manufacturing can allow businesses to do a better job of optimizing for standards (or better). And, it helps recognize problems the moment they start to occur instead of later on when the damage is done and the opportunity for cost savings is lost. This means saving money on raw materials, production, and labor in ways that many are missing out on.
What makes this difficult is finding an easy way to access the data that is meaningful to your shop floor in an affordable way.
Ultimately, most analytics software can’t tell you anything unless they understand your problem; like the scrap rate scenario described above. This is a huge issue for most costly analytics packages, because even when configured properly, they may not know how to best get to the heart of an issue properly or have the expertise to add value throughout the data gathering process.
The problem is that many times organizations do not deeply understand their problem in a way that will allow them to set up analytics for manufacturing either.
ERP and MES packages are too expensive and hard to implement, so they are often intimidating to tackle when a manufacturer may not have 100% clear direction as to what they want to accomplish.
That’s one of the reasons why Mingo has become so attractive to manufacturing businesses. It was designed to collect manufacturing data easily without complex integration of configuration (OEE, availability, how to connect to machines, what data is important).
Manufacturers simply don’t want to explain all of these things to expensive consultants who ultimately will need to be corrected when setting all of this stuff up. Most manufacturers just want the data in real-time so they can make adjustments and save money. This has been expensive and difficult until now.
Small to mid-sized manufacturers are looking for data and the traditional tools don’t allow them to easily collect, analyze, and display this information. Manufacturing analytics offers companies a way to affordably track and report the key metrics that will help increase the bottom line. If you’re ready to get started, think about these 6 questions we ask manufacturers before getting started.
If you’d like to learn more about Mingo, what it does, or how it could help you gather valuable data, watch the Mingo demo. The manufacturers using the tool are getting great results.
Bryan Sapot is a lifelong entrepreneur, speaker, CEO, and founder of Mingo. With more than 24 years of experience in manufacturing technology, Bryan is known for his deep manufacturing industry insights.
Throughout his career, he’s built products and started companies that leveraged technology to solve problems to make the lives of manufacturers easier. Follow Bryan on LinkedIn here.