Predictive Maintenance Expert: Carlos E. Torres Shares His Perspectives and Knowledge
Tell us a bit about yourself. Who is Carlos Torres?
Sure! I'm a globetrotting Salvadoran, born to an Argentine mom and Salvadoran dad. I must confess that I wasn't exactly the best student in school, so when I started studying mechanical engineering, I also decided to pursue a technician degree in automotive mechanics, just in case I didn't finish my engineering degree. Fortunately, I completed both and kept on studying!
These days, I'm a nomad living with my wife between Florida, Spain, Poland, and Switzerland, but we're always exploring other countries. And yes, as you know, I'm the founder of Power-MI, a cloud-based software designed for predictive maintenance management. It's an exciting adventure, I assure you!
Tell us about your professional journey.
Well, my professional journey began while studying mechanical engineering in El Salvador, where I had the opportunity to learn about mechanical vibration mathematics from my teacher, Engineer Orlando Menjivar (Rest in Peace). During my last year in college, I worked as a vibration analyst at SETISA, a service company, and did my thesis on "Predictive Maintenance through Vibration Analysis" under the guidance of Engineer Rodrigo Guerra y Guerra, who was also my boss.
One day, while already working in a steel mill in El Salvador implementing predictive maintenance, I received a call informing me that my thesis had been awarded the best thesis in the country. I thought it was a joke! But it turned out to be true. Thanks to that award, I later got a scholarship to study a master's in mechatronics in Germany. During my master's at the University of Siegen, I worked on a project for diagnosing electric motor bearings based on their electrical current.
In Germany, I worked at SKF and Pruftechnik, always focused on predictive maintenance. Later, Pruftechnik transferred me to Spain, where I was in charge of the subsidiary for Spain and Portugal, also serving clients in Latin America.
During my managerial phase, I completed executive programs at INSEAD in France and Harvard Business School; I hold alumni status from both.
Finally, I left Pruftechnik and embarked on the exciting adventure of founding Power-MI. And here we are!
How was it for you to decide to start Power-MI?
Turns out, Power-MI was a project I had proposed twice to the company where I worked. However, both times, I was sent back to my office to continue with my usual tasks. What irony! The market was ripe, and the business model was perfect to launch the project.
On the other hand, my current partners in Switzerland were looking for a software development project. So, we decided to join forces and venture into it. And you know what? That company that once rejected Power-MI as a project is now one of our clients, using Power-MI consistently. The world takes many turns!
Tell us about Power-MI. How did the idea of creating this tool for managing predictive maintenance come about?
At first, it all started with a need: I wanted a tool that would allow the analysts under my supervision to create quick and uniform-looking reports. But, as I mentioned, that proposal was rejected.
Then, during my studies at Harvard Business School, I discovered how the business world operates and the crucial role of technology. It was at that moment that I understood the future of predictive maintenance wasn't in hardware but in data and, more importantly, information. So, I came up with the idea of creating a digital platform where analysts and maintainers could collaborate, generating data and information that form the basis of the intelligence driving predictive maintenance. Thus, along with my Power-MI partners, we created catalogs of failures, predictive technologies, asset types, and more information structures.
Today, Power-MI is a comprehensive tool for managing predictive maintenance. It not only allows for report creation but also enables work order management, export to CMMS, route design, analyst calendar generation, predictive maintenance savings calculation, fault statistics viewing, root cause analysis, and much more. And to think it all started with the simple need for faster reporting!
Power-MI is not just a predictive maintenance management tool; you also provide training in specific areas. Tell us a bit about this.
As you can imagine, I talk to maintenance managers from different countries and industrial sectors every day. Most of them are dissatisfied with their predictive maintenance or simply don't have it implemented. It's common to hear cases where they bought an instrument and started measuring with routes, hired an external company for inspections and reports, or tried to implement a predictive maintenance plan but failed in the attempt.
On the other hand, I teach predictive maintenance implementation in postgraduate and master's programs. Instead of exams, I assign my students the task of creating a predictive maintenance plan for a set of equipment in their companies. Surprisingly, many of them have told me that, thanks to this task, they have achieved in a short time what they couldn't do in years in their companies.
This gave birth to Planeo-PdM, an approach that Anglo-Saxons call "learning by doing." It's an agile methodology to design a predictive maintenance plan in less than thirty days. Several companies have successfully implemented it, and the best part is that participants don't forget how to design a predictive maintenance plan, leaving a valuable legacy in their companies. So, everyone wins!
Not only are you the CEO of your own company, but you also have a passion for academic training. How did this passion for sharing your knowledge develop?
Initially, the idea of teaching didn't excite me. My perception changed when I started applying the Harvard Business School case method in my classes. I present students with real cases of companies that have implemented predictive maintenance, and we analyze together the decisions that need to be made throughout the process.
Also, with Planeo-PdM, it's truly rewarding to see how students develop their own predictive maintenance plans, ready to be implemented following an executable roadmap. The dedication of the students and their amazement at how simple the process can be are aspects that I am passionate about in training.
Training should be an opportunity for the student to experience and learn. Unfortunately, in the field of predictive maintenance, training has been almost monopolized by consultants whose main goal is to convey academic content so that, in the end, students end up hiring their consulting services. This situation limits the ability of students to effectively experiment and learn.
How do you balance leading your own company and continuing to teach?
Very simple: make classes part of Power-MI's business model. The participant who designs their predictive maintenance plan with the Planeo-PdM methodology has the option to subsequently use Power-MI. Most of them do, as they realize how straightforward things are when managed agilely.
For you, what is condition-based maintenance?
Many students and clients ask me about the difference between predictive maintenance and condition-based maintenance (CBM). In practical terms for the daily operation of an industrial plant, it's the same. CBM is a term introduced by ISO to clarify that condition-based maintenance is not just a prediction of when an asset will fail but also the identification of corrective measures necessary to prevent failure.
Thus, CBM is not just about creating a report but also, based on the report, generating a work order to correct the potential failure. But the story doesn't end there: after the technical closure of the order, it must be verified that the potential failure no longer exists. If it doesn't exist anymore, we must document the case and calculate the savings from CBM. Then, for catastrophic or repetitive failures, root cause analysis should be conducted to be proactive. This is the CBM procedure.
At the end of the day, the ultimate goal of condition-based maintenance is to save money for the company. If these savings are not demonstrated, the work of the people involved becomes irrelevant.