??Predictive Maintenance [PdM]: Industrial IoT "Killer Application"??

Hello everyone,

Hope you enjoyed Easter break with your family and friends, as I surely did!

Happily back to my weekly newsletter, I want to focus on what I always considered Industrial IoT's low-hanging-fruit application: Predictive Maintenance [PdM]:

Back in October 2017, when with my company VINCI Digital | IIoT + AI / GenAI Strategic Advisory ?? I presented at IOT Solutions World Congress in Barcelona, Spain

"...a renewables use case / case study of a consistent and fully scalable end-to-end Industrial IoT solution, to be deployed initially as a pilot on an 11 MW solar farm site of Enel Green Power in Brazil. A customized full-stack IIoT solution, focusing mainly on predictive maintenance advanced analytics application (AI / ML-based algorithms), in order to boost both solar panels productivity and overall O&M efficiency gains."

...I wasn't using a crystal ball, but - as a former strategic consultant and industrial executive - I already knew that costs cutting and productivity increase of operations / supply chains with predictive maintenance applications would be the focus of every industrial company, trying to validate Industrial IoT solutions' effectiveness and efficiency to get their first relevant and scalable RESULTS, in months, not years!

Now in 2023, Predictive Maintenance's name of the game is:

1. Seamless systems integration with current OT / IT infrastructure [legacy systems, ERPs, CRMs, etc.], which is still a barrier to overcome in terms of interoperability and skilled / available service providers;

2. Vertical / use case driven AI / ML advanced algorithms, better if trained on a cloud data lake [private and/or public], compiled not only with traditional CNC/PLC, SCADA, MES data inputs, but eventually through advanced Industrial IoT sensors - possibly in retrofit, for a quick set-up - measuring in real-time key variables, such as vibration, temperature, noise, electromagnetic field, etc.;

3. Transition from predictive to state-of-the-art Prescriptive Maintenance solutions, where not only optimization is the main driver, but also potential autonomous actions - triggered by real-time alarms, and possibly ERP integrated - will together results in significant time saving, costs cutting, and revenues increase.

? Prescriptive Maintenance [PsM] takes PdM a step further by prescribing corrective actions for deteriorating conditions and including them in the alerts.

? PsM is a newer concept just gaining ground that is made possible by machine learning [ML], artificial intelligence [AI], and the internet of things [IoT] = AIoT.

? Prescriptive Maintenance [PsM] uses custom-built prescriptive algorithms to process the potential outcome of multiple scenarios.

? Prescriptive analytics so is typically used to identify the best option in a broad field of choices.

? It uses advanced modeling based on?artificial intelligence [AI] and machine learning [ML]?or deep learning [DL] to examine the potential consequences of different courses of action, and then recommend the best path to take.?

? The result??A list of possible solutions, ranked based on predetermined factors you defined while building the algorithm.

4. As a result, several competitive startups were already offering in 2017 Predictive [PdM] and Prescriptive [PsM] maintenance solutions, each one focusing a specific vertical, such as

??Manufacturing

??Transportation

??Energy & Power Generation

??Oil & Gas

and a different application / use case

??Prescriptive Maintenance [PsM]

??Predictive Maintenance [PdM]

??Condition-Based Maintenance [CBM]

as you can see in the following PdM / PsM Startups Market Map:

Nowadays, the PdM / PsM startups market is still growing, very competitive, and consolidating, already creating the first Scale-Ups and Unicorns (see picture below). Some barriers are still to overcome - but not impossible ones! - such as:

??Limited availability and ownership of "good" data: sh*t in ? sh*t out

??Security and privacy ? lack of trust in systems

??PoC / Pilots and traction ? Commercial Deployment / Scale

Hope you liked it, and please find below more Insights & Perspectives, and Predictive & Prescriptive Maintenance case studies.

See you next week!

Ciao, Fabio

---------------

INSIGHTS & PERSPECTIVES

Predictive Maintenance – Understanding the Value of Truly Predictive Asset Optimization | IoT Analytics Webinar - By Knud Lasse Lueth | March 30, 2023

State of Industry 4.0 in 2023

??As you can see from the chart above, IoT spending in Manufacturing is forecasted at $132.9 billion in 2023, with an estimated growth (y/y) of 19.2% [$111.5 billion in 2002!]

??Operational improvement is the biggest goal for manufacturers, and Predictive Maintenance [PdM] influences all top 5 KPIs:

? Increase in overall equipment effectiveness [OEE]

? Increase in labor efficiency

? Increase in output

? Decrease in costs

? Increase in quality

What Predictive Maintenance does:

??OEE: Leads to less downtime which results in higher asset availability ? higher OEE

??Labor efficiency: Leads to reduction in regular maintenance which frees up labor ?better efficiency

??Output: Less downtime means more time to produce ? higher output

??Costs: Fewer replacement parts, less personnel, lower Capex (due to longer asset life) ? Lower costs

??Quality: Less scrap due to unforeseen production stoppage ? higher quality

How Predictive Maintenance works in practice - common example

Hybrid Modeling: Improving prediction accuracy by leveraging more/different data sources

Source / Webinar Video: https://youtu.be/OfDPijTofhM

#iot #ai #aiot #manufacturing #predictivemaintenance #pdm #operationalimprovement #oee #efficiency #output #costs #quality

----------

Combine Predictive and Prescriptive Analytics for Better Decision Making | Via Gartner | September 2022

To help shape the future of the organization and get improved value outcomes from their decision making, data and analytics leaders must know when and how to combine predictive and prescriptive analytics to gain a competitive advantage.

Select the Optimal Mix of Predictive and Prescriptive Analytics

??The decision of which predictive and prescriptive analytics techniques to use and how to combine them depends on the nature of the problem and the complexity of its potential solutions.

Each type of analytics has distinct benefits

? Predictive?analytics?can be used to forecast a series of outcomes over time. This type can also be used to highlight uncertainties related to multiple possible outcomes, relying on techniques such as forecasting, regression, multivariate analysis, pattern matching or ML. Data and analytics leaders should focus on answering the question, “What is likely to happen?”

? Prescriptive analytics?calculates the best way to achieve or influence the outcome and to drive action, relying on techniques such as neural networks, optimization or recommendation engines, answering the question, “What should be done?”

??When combined with predictive analytics, prescriptive analytics naturally draws on and extends predictive insights. For example, Bayesian analysis (predictive) can combine with optimization analysis (prescriptive). This research examines five common analytics techniques: three predictive and two prescriptive (see figure below).

The nature of the problem will guide the choice of whether to use prediction, forecasting or simulation for the predictive analysis component. The complexity of the solution will guide the choice of whether to use rules or optimization for the prescriptive analysis component.

Case Study 1: Predictions and Rules

A hospital needed a way to monitor and reduce surgical site infections. The goal was to maximize the value of care provided by the hospital to improve patient outcomes, reduce clinical variations in care provided and reduce the cost of hospital care.

Case Study 2: Forecasting and Optimization

A truck and trailer parts and services company managed multiple distribution centers and operated more than 260 locations across the U.S. — delivering vital spare parts for heavy-duty vehicles that have suffered mechanical issues and need repair. Customers relied on the company’s differentiated ability to deliver replacement parts quickly at the right price.

Case Study 3: Simulation and Optimization

A financial services company that provides auto financing to more than 4 million customers needed to reduce auto loan delinquencies and repossessions. The organization wanted to reach this goal by providing payment options that are profitable to the business, but taking into consideration individualized customer preferences, avoiding repossessions and credit impact, while profitably growing the lending portfolio.

Case Study 4: Optimization Only

A global snack-food manufacturer needed to coordinate efforts between production of potato chips and allocation of production output to multiple distribution centers. Production had been assigned only on cost of transportation. The company’s objective was to balance cost of production, throughput, transportation and cost of inputs.

Source: https://gtnr.it/3KYHLJc

#industrial #iot #iiot #ai #aiot #ml #predictivemaintenance #pdm #predictiveanalytics #prescriptivemaintenance #psm #prescriptiveanalytics

---------------

CASE STUDIES

???????Amazon Web Services (AWS) ????????? ???? ???? H???????Amazon ??????????????????????? ?????????????? ???????????? ???????????????? ???? ????% ?????? |?Via?VentureBeat ?| February 2023

???????? ?????? ??????????, ???????????? ?????????????????? ???????? ???????????? ???????? ???? — ?????? ???????? ???????? ???????????????????????? ?????? — ?????????????????? ????????????????.

??But it doesn’t happen by magic, of course. Instead, packages at the company’s hundreds of fulfillment centers traverse miles of conveyor and sorter systems every day, so?Amazon ?needs its equipment to operate reliably if it hopes to deliver packages to customers quickly.

??To take on this challenge, the retail leader has announced it uses Amazon Monitron, an end-to-end?machine learning (ML) ?system to detect abnormal behavior in industrial machinery — that launched in?December 2020 ?— to provide predictive maintenance.

Monitron includes:

• Sensors to capture vibration and temperature data.
• A gateway to securely transfer data to the AWS Cloud.
• A service that analyzes the data for abnormal machine patterns using machine learning.
• A companion mobile app to set up the devices and track potential failures in your machinery.

??As a result, 亚马逊 has reduced unplanned downtime at the fulfillment centers by nearly 70%, which helps deliver more customer orders on time.

Amazon Monitron solves real-world industrial problems

“One of the key things that Amazon does is they take technologies like machine learning and they apply them at scale to solve real world problems,” Vasi Philomin , VP of AI services at?AWS , told VentureBeat . “That’s really what drew me to this company in the first place.”

??According to 亚马逊 , up to 80 engineers are responsible for maintaining the equipment at each fulfillment center. Before implementing Amazon Monitron, technicians walked around the site, taking readings and manually analyzing the measurements to determine the condition of the equipment, including ultrasound, thermography and oil analysis.

Source:?https://bit.ly/3lnAFmR

#iot ?#ml ?#aws ?#amazon ?#fulfilment ?#downtime ?#fastdelivery

----------

‘Predictive-Maintenance’ Tech Is Taking Off as Manufacturers Seek More Efficiency | Via The Wall Street Journal / CIO Journal | September 2022

百事 , 高露洁 and other firms are populating their plants with sensors from AI startup Augury to ‘listen’ for machinery problems. And other up-and-coming ‘machine-health tech’ firms [such as? C3 AI , DataProphet , and Senseye - A Siemens business ] are offering similar wares

Startups that make technology designed to predict industrial equipment failures before they happen are seeing a surge in demand, as strained supply chains prompt manufacturers to squeeze more efficiency out of production lines, startup founders and analysts say.

百事

?? Anna Farberov , General Manager of?PepsiCo?Labs, the technology venture arm of PepsiCo Inc., said that over the past year so-called predictive-maintenance systems at four Frito-Lay plants reduced unexpected breakdowns, interruptions and incremental costs for replacement parts, among other benefits.

“At 百事 , we're always looking to improve our operational excellence. And specifically, we wanted to reduce our unplanned downtime. We had a very clear business target that we had to achieve, and we scouted for a few solutions, big companies, small companies. We did an objective side-by-side assessment. We tested a few actually. Augury came as the best solution and the most fitting one for the 百事 operations.” - Anna Farberov , General Manager, PepsiCo Labs

??Developed by New York-based startup Augury Inc., the technology has helped Frito-Lay add some 4,000 hours a year of manufacturing capacity—the equivalent of several million pounds of snacks coming off the production line and shipped to store shelves, Ms. Farberov said.

“There was improved performance in terms of reduced unexpected breakdowns, interruptions, and incremental costs for replacement parts on the monitored assets. In addition to this enhanced productivity, importantly, it has enabled a better work experience for our associates so they can focus on delivering quality products for consumers.” - Clark Michael , Supply Chain Director, PepsiCo

COLGATE-PALMOLIVE

?? warren pruitt , Vice President of Global Engineering at 高露洁 , said the company turned to predictive-maintenance tools in a bid to improve machine performance and reduce machine downtime. Previously, the company relied on preventive and calendar-based maintenance to manage equipment, he said.

?? 高露洁 has put Augury ’s platform to work in all of its North America plants, as well as many in Europe, Latin America and Asia, he said.

“Our predictive-maintenance program also upskills our workforce, giving our employees the bandwidth to look at the big picture and consider how to employ new technologies and initiatives to continuously improve our operation,” Mr. Pruitt said.

Source: https://on.wsj.com/43x9Rm6

#industrial #iot #iiot #ai #aiot #ml #predictivemaintenance #pdm #startups #operationalexcellence #downtime #productivity

---------------

Bio:?Fabio Bottacci ?is a relationship builder, creative problem solver and strategic thinker. Senior industrial executive, he acquired a solid background in large multinational organizations across Brazil, US, and Western Europe. He is known for his ability to deliver results despite ambiguity and obstacles, to build bridges between people and to manage conflict and negotiations.

He began his career at?Accenture Italia , strategy practice, while attending MBA courses. He then moved to Brazil, where he consistently proved, during more than 20 years of professional experience, strong client's network, industry knowledge and business development expertise in the oil and gas, automotive and energy / utilities verticals.

Since 2015, he has been the founder & CEO of?VINCI Digital - Industrial IoT Strategic Advisory , being recognized internationally as a thought leader by well-known organization, such as the?Wor ld Economic Forum ,?IoT Solution World Congress ,?BNDES (the Brazilian Federal Development Bank), etc., and helping startups, SME, and corporations to thrive within the actual digital transformation environment, by increasing productivity, developing new business models, and - most important of all - delivering actual results / ROI in months, not years.