12
THE JOURNEY The Manufacturing Analytics Journey BY BILL BITHER, CO-FOUNDER AND CEO OF MACHINEMETRICS AND GRAHAM IMMERMAN, VP OF MARKETING, MACHINEMETRICS

THE JOURNEY Ready... · Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production

  • Upload
    others

  • View
    1

  • Download
    0

Embed Size (px)

Citation preview

Page 1: THE JOURNEY Ready... · Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production

THE JOURNEYThe Manufacturing Analytics Journey

B Y B I L L B I T H E R , C O - F O U N D E R A N D C E O O F M A C H I N E M E T R I C SA N D G R A H A M I M M E R M A N , V P O F M A R K E T I N G , M A C H I N E M E T R I C S

Page 2: THE JOURNEY Ready... · Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production

T H E M A N U F A C T U R I N G A N A LY T I C S J O U R N E Y

Contents

—PA RT ONE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 03

The Manufacturing Analytics Journey . . . . . . . . . . . . . . . . . . . . 03

What Is Manufacturing Analytics? . . . . . . . . . . . . . . . . . . . . . . . 03

What Are The Advantages of Leveraging Manufacturing Analytics? . . . . 04

—PA RT T WO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 05

From Descriptive to Prescriptive . . . . . . . . . . . . . . . . . . . . . . . . 05

Descriptive Analytics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 06

Diagnostic Analytics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 08

Preventative Analytics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 09

Prescriptive Analytics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

—PA R T THREE  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

How Data on Production and Quality Combine to Create Useful Analytics 11

Going from descriptive to predictive analytics . . . . . . . . . . . . . . . . 11

Page 3: THE JOURNEY Ready... · Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production

0 3

T H E M A N U F A C T U R I N G A N A LY T I C S J O U R N E Y

Part One

THE M A NUFAC T URING A N A LY TIC S JOURNE Y In the past, if you wanted to know how a machine was performing on a shop floor, you would have had to put pencil to paper and note the throughput, the error rate, the quality, all manually . You would have to look at the quality of the inputs and the available supply of the raw materials . You would have had to look at the throughput over time to see if there was a consistency to quality or errors that needed to be investigated .

In other words, you were looking at potentially weeks of manual effort to figure out why and how a manufacturing process was going awry . That, in and of itself, isn’t a good use of anyone’s time and is a pretty inefficient way to become more profitable.

Then came Industrial IoT (IIoT) and the ability to monitor machines directly, providing quantities of data from the machine, as well as the operator, that can be analyzed in order to improve both production and quality . According to research, IIoT has the ability to push the manufacturing industry’s productivity rate and generate approximately $1 .2 trillion for manufacturers worldwide .

W H AT I S M A NUFAC T URING A N A LY TIC S? Analytics is essentially the collection and manipulation of large quantities of data to reveal insights . In manufacturing, the large quantities of data collected come from production equipment, logistics and supply chain management, and shop floor operators. These data from different sources and processes are collected and reformatted as easy to understand metrics, to reveal where there are issues with performance or output quality .

Thus, manufacturing analytics goes beyond the actual collection of the data to include the formulation of insights that can be used at every level of the organization . The advancement of tools and software in this area means that the process of collecting data is no longer manual and the analysis is centrally available, in real-time . This ensures analytical insight is available for everyone from the shop floor manager to the CEO to review and, most importantly, take action on .

Page 4: THE JOURNEY Ready... · Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production

0 4

T H E M A N U F A C T U R I N G A N A LY T I C S J O U R N E Y

WHAT ARE THE ADVANTAGES OF LEVER AGING MANUFACTURING ANALY TICS?

“Increased revenue (33.1%), increased customer satisfaction (22.1%) and increased product quality (11%) are the top three benefits of Industrial Analytics.” (Source)

THERE ARE MANY ADVANTAGES TO LEVER AGING REAL-TIME ANALY TICS:

Customer satisfaction The ability to discover performance and / or quality control issues in every individual order before they become problematic, leads to better customer satisfaction .

Availability of real-time data Rather than waiting for a problem, the availability of real-time data that is formatted into metrics and insights constructed for specific users, a problem with a machine can be proactively discovered, even predicted based on historical data and current usage .

Reduction in errors Tracking the Mean Time Between Failures (MTBF), leads to an optimized manufacturing process . With MTBF, errors that may cause items to be scrapped will be detected early . Early detection means less expenses for the manufacturer .

Reduction in unplanned downtime The real-time data also allows operators to get an immediate view of the performance of any given machine; a minute lag in production or issues that may affect quality can be picked up in real-time . This ensures the needed maintenance or repairs can be scheduled to forestall further damage to machines . Thus, instead of operators existing in reactive mode, they’ll be able to be proactive in handling the maintenance of machines . Eventually, with the application of artificial intelligence to machine data and data analytics, AI will enable machines to fix themselves without human supervision .

Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production efficiency levels. The reduction in scraps, downtime, and maximizing operator schedules lead to huge recurrent savings for manufacturers.  

Page 5: THE JOURNEY Ready... · Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production

0 5

T H E M A N U F A C T U R I N G A N A LY T I C S J O U R N E Y

Part Two

THE MANUFACTURING ANALY TICS JOURNEY FROM DESCRIPTIVE TO PRESCRIPTIVE

• Descriptive Analytics, which use data aggregation and data mining to provide insight into the past . Descriptive analytics answers the question; “What has happened in the past” and

“What is happening now”

• Diagnostic Analytics: Diagnostic analytics attempts to understand and answer the question, “Why it happened?”

• Predictive Analytics, which use statistical models and forecasts techniques to understand the future . Predictive analysis answers the question; “What could happen?”

• Prescriptive Analytics, which use optimization and simulation algorithms to advise on possible outcomes . Predictive asks the question “What should we do?”

D E S C R I P T I V E

What is happening?

P R E D I C T I V E

What is likely to happen?

D I A G N O S T I C

Why is it happening?

P R E S C R I P T I V E

What should I do about it?

Page 6: THE JOURNEY Ready... · Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production

0 6

T H E M A N U F A C T U R I N G A N A LY T I C S J O U R N E Y

DE SCRIP TI V E A N A LY TIC S

Descriptive analytics is the interpretation of historical data to understand the performance and changes that have occurred in a business . If you want to know what happened, use descriptive analytics and until recently, this is how most companies used data—to see what had happened in the past . Descriptive analytics provides insight into the past and showcases the effects of the decisions that were taken. The analytics can be customized to fit any time frame or set of customers . If you want to know what has happened in the last 30 days or year with only a certain demographic of customers, descriptive analytics can get you accurate answers .

In customer experience, descriptive analytics can be used to track things like total tickets and resolutions . They can be especially helpful in tracking trends to help plan for the future . If the number of customer calls increased greatly around the holidays or when a new product launched, you can plan for increased call volume around those events in the future . Descriptive analytics can also help provide a better understanding of customers by tracking their shopping and contacting preferences .

DESCRIPTIVE ANALY TICS USE CASES FOR MANUFACTURING

In manufacturing, descriptive analytics has multiple use cases . These cases include; evaluating past production cycles, providing benchmark data, optimizing machine performance, and developing a foundation for diagnostic and predictive analytics .

To understand these use cases, the example of a machine shop producing metal components paints a clearer picture . In this scenario, the machine shop collected machine and production data through the first quarter of 2019 and intends to use the data for descriptive analytics . With the collected data, the machine shop successfully:

• Evaluated 2019’s Productivity Levels With descriptive analytics, the machine shop was able to highlight which month was the most productive and what made it so . It could also track downtime and the causes from the historical data .

• Benchmark Data The collected data and the evaluation done provides benchmark data highlighting the most productive cycle through the months and why productivity was optimized .

• Develop a Foundation for Predictive Analytics To effectively plan for the future, knowledge of the past is required . The benchmark data descriptive analytics provide will be used by the machine shop to estimate machine production capacity, the effects of increased demand and downtime etc .

Descriptive analytics then puts the machine shop in a better position to plan for optimized productivity throughout 2020 . It also provides manufacturers with insight into the important data that was overlooked so solutions can be put in place to capture them in the future .

MachineMetrics Operator View allows operators to add human-context to machine data with a touch screen interface mounted right at the machine tool. Categorize downtime, reject a part, and start/stop jobs to manage and record quality data.

Page 7: THE JOURNEY Ready... · Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production

0 7

T H E M A N U F A C T U R I N G A N A LY T I C S J O U R N E Y

THE IMPORTANCE OF BENCHMARK DATA FOR DESCRIPTIVE ANALY TICS

No manufacturing service or operation exists in a vacuum . Today, the manufacturing industry happens to be one of the most competitive industries out there . Thus, to compete manufacturers routinely compare their productivity metrics with those of the competition .

Benchmark data refers to the optimized industry-wide data of a machine needed for these comparisons . Thus, with benchmark data, you can evaluate both current and older machine

data against the benchmark for your machine to determine or discover production shortfalls .

Using benchmark data as a descriptive analytics tool, manufacturers can answer questions such as: What do we do to optimize machine performance to meet industry standards? What are we doing wrong that makes us lag behind the competition?

Source: The ‘2019 State Of The Industry CNC Machining’ report, MachineMetrics.

UTILIZATION HEATMAP, ALL COMPANIES

Page 8: THE JOURNEY Ready... · Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production

0 8

T H E M A N U F A C T U R I N G A N A LY T I C S J O U R N E Y

DIAGNOSTIC ANALY TICS

Diagnostic analytics involves the analysis of historical data and real-time data with the aim of answering the question ‘Why did this happen and why is it happening’ . In contrast to descriptive analytics, diagnostic analytics is less focused on what has occurred but rather focused on why something happened . In general, this analytical process involves looking at the processes and causes, instead of the result . Here is an example of diagnostic analytics “A machine is working optimally for longer durations and the likely reason is due to the fluid change and servicing activities that were done a week ago .”

Thus, diagnostic analytics show that the fluid change is why the machine is functioning optimally . On the other hand, descriptive analytics will record that timely fluid changes lead to increased performance . Also, take note that descriptive analytics cannot provide an answer to important questions such as “How can we avoid this problem” or “How can we duplicate this solution?” These are covered by diagnostic analytics .

DI AGNOS TIC A N A LY TIC S USE CA SE S FOR M A NUFAC T URING

The manufacturing industry relies heavily on diagnostic analytics to ensure production cycles are efficiently run. The application of diagnostic analytics covers asset maintenance, dealing with downtime, and also providing a foundation for predictive and prescriptive analytics .

Using the machine shop as an example, the machine data it collected during a morning shift showed that its machines broke down regularly during the 3rd shift . Using diagnostic analytics to delve deeper into the data collected during the durations the alarm rang out, it was discovered that a program was changed the previous day. These program changes were identified as the underlining cause of the failures .

To eliminate the breakdown that occurred during the 3rd shift, the programmer reviewed the codes from the previous day, debugged it, and correctly deployed it . This successfully solved the issue and stopped the alarm from ringing . Thus, with diagnostic analytics the machine shop successfully;

• Discovered Causation Correlating machine data with historical data, the shop floor carried out diagnostic analyses to know the cause of its problems .

• Developed Solutions The analytics done also provided a basis for initiating a solution to ease the problem and stop it from re-occurring in the future .

• Built a Foundation for Predictive and Prescriptive Analytics Successfully understood why this is happening ensures the solution found can be applied in the future before a breakdown occurs .

View and export real-time machine data and alarms as time-series data and charts to help diagnose and resolve problems.

Page 9: THE JOURNEY Ready... · Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production

0 9

T H E M A N U F A C T U R I N G A N A LY T I C S J O U R N E Y

PREDIC TI V E A N A LY TIC S

Predictive analytics involves the analysis of historical and real-time data to provide insight into what could happen in the future with high-levels of accuracy . With predictive analytics, the future is no longer a blank plate and this is made possible through the use of data analytics, artificial intelligence, simulation, and machine learning .

AI-powered platforms employ machine learning and anomaly detection to analyze large data sets to create forecasts for specific months, days or years. The large data sets are collected from business operations and assets . The collected data serve as the basis for predictive analytics .

PREDIC TI V E A N A LY TIC S USE CA SE S FOR M A NUFAC T URING

In manufacturing, predictive analytics has become an important concept because of its ability to help manufacturers plan for the future . The data sets collected in manufacturing to aid predictive analytical initiatives, include real-time machine data, inventory data, data from descriptive and diagnostic analytics etc . Once collected, predictive analytics can be initiated and using the machine shop as an

example, predictive analytics can be used to drive production scheduling policies, predictive maintenance, optimize machine performance, and enhance productivity levels .

• Develop Future Schedules With increased demand comes the need for increased production activities and only a few shop floors can switch up their production capacity without adequate planning . Thus, predicting future demand enables the creation of schedules that take into account the expected change in consumer consumption rate .

• Predictive Maintenance Equipment breakdown leads to downtime and a less motivated workforce . With predictive analytics, unplanned downtime becomes planned downtime which gives operators scheduled time-offs for other planned activities . Patterns in the data can indicate that a component of the machine is beginning to fail . Predicting when that component might fail means maintenance teams can plan to preventatively fix the problem before it causes unplanned downtime .

• Enhance Productivity Levels Predictive analytics provide a basis for advanced capacity planning which enables manufacturers to optimize machine efficiency and productivity levels .

With MachineMetrics Edge platform, you can easily collect and analyze low and high-frequency machine data to quickly develop and deploy predictive analytics algorithms.

Page 10: THE JOURNEY Ready... · Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production

10

T H E M A N U F A C T U R I N G A N A LY T I C S J O U R N E Y

PRE SCRIP TI V E A N A LY TIC S

One of the newest types of data analytics is prescriptive analytics, which takes descriptive, diagnostic, and predictive data and uses it to make recommendations for future actions and improvements . Prescriptive analytics give advice on potential outcomes and recommend what will happen if those outcomes are reached . By pinpointing the best outcomes and making recommendations, companies can focus on delivering amazing customer experiences . Prescriptive analytics help companies make decisions that reflect customers’ needs and changing trends .

Some credit card and insurance companies use prescriptive analytics to analyze past factors like purchase history and credit score, to predict how a customer will behave in the future and what actions the company can take . Prescriptive analytics is action-based and helps keep the company ahead of trends to make smart, future-focused decisions .

Prescriptive analytics allows lesser skilled workers to be provided instructions that would normally require a more experienced higher skilled worker with tribal knowledge . Advances in prescriptive analytics will improve the skills gap in many industries, including manufacturing .

PRE SCRIP TI V E A N A LY TIC S USE CA SE S FOR M A NUFAC T URING

In manufacturing, prescriptive analytics can be applied in diverse ways to deliver an optimized production environment and to create customer behavioral models that influence decision making processes. A manufacturer can use descriptive data to develop multiple products which will appeal to customers according to age demographics or location .

This expansion will require an increase in production capacity, the need for an updated inventory, and advanced planning . Using the machine shop as an example, expanding operations to include the machining of complex parts will require different materials and tools . With prescriptive analytics, the machine shop can devise multiple solutions which take into consideration capital and equipment constraints . The machine shop stakeholders can then choose the most viable option for expanding their operations .

In one example, when a machine alarms out, we can deliver instructions to the operator on how to fix the problem when it occurs . In another example, we can deliver a message to the CNC programmer in real-time that a program change caused a quality issue, giving the information to the programmer that recent change to the program caused an inadvertent issue . A DA P TI V E C ONTROL

Prescriptive analytics typically involves the human in the loop to make pro-active change . With control over the machine, changes can be made directly to the machine to improve throughput and avoid failures . These features are often built into the machine itself, but with the advent of machine learning, and the benefit of newer data sets, after-market features can be developed with this data . For example, machine learning algorithms can be developed to detect cutting tool wear, and instruct the machine to change the tool - preventing unplanned downtime, improving quality, and improving tool life .

Create and manage preventative maintenance schedules in MachineMetrics that are tied to calendar time, usage time, or initiated from machine conditions. Assign maintenance tasks through a workflow.

Page 11: THE JOURNEY Ready... · Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production

11

T H E M A N U F A C T U R I N G A N A LY T I C S J O U R N E Y

Part Three

HOW DATA ON PRODUCTION AND QUALIT Y COMBINE TO CREATE USEFUL ANALY TICS

“69% of decision-makers believe Industrial Analytics will be crucial for business success in 2020, with 15% considering it crucial today.”(Source)

Production and quality are undeniably interrelated . A gain in one that results in a loss in the other is not moving the manufacturing process, or its associated costs, forward . In the past, data on performance was often looked at separately from quality . Whether a machine—or their operators—were lagging during the overnight shift set up on a big order was seen as distinct from the quality of the output .

The theory being that the customer would be okay with a longer production time on their order as long as the quality was there . But that’s no longer the case . Customers the world over are expecting more and rightfully so . To achieve a better level of satisfaction, the data on performance AND quality must be combined to achieve industry or organizational level markers . These metrics are the difference between keeping a client happy and simply keeping them .

GOING FROM DE SCRIP TI V E TO PREDIC TI V E A N A LY TIC S

The bottom line in manufacturing analytics is to go beyond the simple collection and display of data (descriptive) to leveraging it at a more

granular level to be able to predict issues on the shop floor and save money while improving output, in the process . Taking historical data, for example, and estimating what isn’t known, such as future demand, the statistical possibility of machine failure and so on will help everyone from the operator on the shop floor to the higher-level decision-makers to see where production is at or likely to be, at any given moment .

MachineMetrics provide diverse tools that enable manufacturers to collect shop floor data and go from descriptive to predictive analytics while reaping the accrued benefits. As with any prediction or estimate, there is no guarantee of accuracy, but it’s far more likely to be valid when enough data is used to support the analytics, than human predictions .

All in all, the manufacturing industry is seeing that data analytics aren’t just about ‘keeping track of things’ but about increasing revenue, reducing costs and maintaining a level of customer service that will keep it relevant in the years to come .

Page 12: THE JOURNEY Ready... · Reducing manufacturing costs A data-driven manufacturing process optimizes machine utilization, enables predictive maintenance, and improves overall production

CONTACT US 413-341-5747 info@machinemetrics .com machinemetrics .com

ABOUT MACHINEMETRICS MachineMetrics is manufacturing’s first Industrial IoT Platform for Machines. We transform analytics into action through universal machine connectivity, cloud data Infrastructure, and communication workflows that optimize machine operation . Right now, hundreds of manufacturers have connected thousands of machines to MachineMetrics across global factories . Our platform is enabling these companies to deliver the right information to the right person at the right time to improve their machine performance and productivity, increase their capacity utilization and ultimately win more business to remain globally competitive .