Deep dive - Predictive Maintenance in the Food and Beverage industry

Trend Detection Podcast ·

Show notes

Welcome to the Trend Detection podcast, brought to you by Senseye Predictive Maintenance – the platform, powered by Siemens, which enables predictive maintenance at scale across all of your assets, across all of your plants.

For this episode, Kelli Case and Emily Johnson discuss the ins and outs of predictive maintenance in the food and beverage industry.

Please subscribe via your selected podcast provider to be notified about future episodes.

You can find out more about how Senseye Predictive Maintenance can reduce unplanned downtime and contribute towards improved sustainability within your manufacturing plants, by visiting:

www.siemens.com/senseye-predictive-maintenance

Connect with Kelli Case:

https://www.linkedin.com/in/anshill

Transcript

Welcome to the trend detection podcast powered by Siemens. What do you think the current market appetite is for predicted maintenance in the food and beverage industry? I believe there's a great deal of opportunity. Many of the companies are trying to go through sustainability efforts, so both from environmental impact, but also in the way that they're sourcing ingredients and and those types of things. So predictive maintenance from a sustainability perspective is, is a great addition. If your equipment's not performing and behaving the way that it's supposed to and up to, you know, standards, then you need to know about that, and you can go fix a problem before an environmental situation happens and also dovetails into the energy consumption, so bad actors consume more energy. So being able to to catch those issues ahead of time is, is also a great opportunity. So that that's one area of focus. Two, you know, high demands require that you have a good capacity to meet your customer demands from a. Product perspective, so you can't do that if your equipment is consistently down or you're having problems with unplanned downtime. Uh, and, and third, bad actors also impact the quality of the products, um, that, that you're producing, so over baking something. Um, you know, for example, can, can have quality issues that then turn around and hurt you in, in the marketplace. So that brand recognition, being able to deliver the, the way that you want, um, all of those can be, uh, encouraged and supported using, uh, predictive maintenance. Yeah, no, 100% that's that. No, that's a brilliant answer, thank you so much. That's a like a really good summary. It's really interesting to see kind of the appetite in different industries, um, because, yeah, I think obviously each industry has its own specific needs, so it's really interesting to see how, um, the, the platform itself can cater to, you know, each industry's um, and yet their, their needs at the moment. So thank you for answering that. Um, so then my second question is, um, Probably, I mean, probably similar to what I've just asked, but um what do you think for um the for the food and beverage industry, what do you think are the are the benefits that they, you know, hope to gain from using predictive maintenance, what would you say are the specific benefits, um, if, if you give a a couple of examples of what they could hope to gain from um deploying predictive maintenance. So in, in the food and beverage industry, when you're plagued with problems of unplanned downtime, it can have a significant impact on the materials, as you can imagine, once you start the process, you may have a window of time to be able to to produce the product, otherwise you get degradation in the ingredients. You can have situations where unplanned downtime, if you're in the middle of that. Production, depending on the material that that you're working with, can gum up the machines which then requires additional cleaning and sanitation to get because you are working in a sanitized environment many times in the food and beverage industry plus you have the additional cost potentially of either having a third party come and pick up the waste, you know, of, of those. Ingredients, it can impact the stock of those ingredients that you may have that again can, you know, have a negative impact on capacity and the ability to meet your customers' orders and and demands. So those are just a few small components of, you know, the potential of, of where predictive maintenance if you have the ability to say in 2 weeks we need to take this line down. That gives you the ability to schedule the maintenance. It, it gives you the ability to schedule within the production when is the best time to, you know, take the system down. Uh, it also allows you to start understanding that the inventory of spare parts and what you should really have, um, on the shelf, what can be delivered, you know, overnight versus what has delivery of, of weeks or months, uh, lead time, especially in, in the supply chain environment that we're in, um, that can be very important. Um, and then one other aspect is, is predictive maintenance really helps you have a, a better understanding of what maintenance needs to be performed and when, uh, which helps you have a, a better preventative maintenance schedule cause you still need to change oil on, on a regular basis. But if you can do it with data-driven information. Maybe you can go from 3 months to every 4 months, so that reduces the the safety risk to your employees. The fewer times you have to open up a piece of equipment or walk on a scaffold, the better it is for, for your maintenance staff. Uh, it reduces the consumables. So if you're changing oil, you know, 3 times a year versus 4 times, um, it, it's a, you know, a money saving, uh, you know, financial stability for, for a company from, from that perspective. Yeah, no, 100%, that, that's a great answer, thank you so much, and Have you found kind of um with your, your clients and your your customers, um, have they been more susceptible to wanting to do a pilot run or have you felt like they're a bit resistant to um adopting a scalable approach, so are they wanting to kind of just test it on a few machines and test it at that, or do you feel like, um, some of them have been a little bit more, um, I don't know, understanding and want to kind of, deploy the platform in a scalable way, um, be interesting to just hear your take on that. I, I think oftentimes they start with the idea that they want a scalable system, um, especially some of the larger corporations that have plants, you know, around the United States, around the Americas, and, and, you know, all the way around the globe, having a solution that that can scale easily without additional staff, without having to hire and or train data scientists to. Be able to deploy a solution like this is, it is really important to, to organizations. But with that being said, the food and beverage industry is, I think, similar to, to a lot of other industries that, you know, show me, prove it to me, and, and, and then we'll scale. Um, some, some companies are, are more forward thinking in adopting technology and embracing, you know, the industry 4.0, 5. You know, um, concepts and, and those types of things. So, um, it is somewhat dependent on a company's culture and their openness to embrace and and really leverage technology. But even within those kind of cultures, the, let's start with, with a plant. Let's get the processes and the, you know, the information kind of settled, uh, and, and then we'll take it to, uh, other, other locations. So yeah. Yeah, no, thank you again, that's a great answer. Um. So, next question, um, so just from your perspective, what do you think a client needs to do to ensure a successful predictive maintenance deployment? So would you give advice on kind of making sure that the data is prepped, for example, just removing data silos, cleansing, cleansing data. Have you got any kind of secret tips or advice that would be good for, um, a, a prospective customer or client? Uh, oftentimes it, it starts with, you know, take it, take it in a phased approach, start, you know, on, on the smaller sail side. Uh, don't try to, you know, boil the ocean, you know, as the, the old saying goes, um, because it It, it is a large project to take on. And so some, some of the advice is find a champion within your organization that believes in the concept, believes in, in the technology and what it can bring to an organization so that they can promote the benefits as they start to come in when those successes happen. So, you know, proof that the system brings the value. Uh, that, that we've talked about to, to start with. 2 is, is education. You know, people are not comfortable with systems that, that they don't know why they're using it. They don't know how to use it. They'll revert back to the old ways because they have a job to get done. Um, we hope to make that job better, more efficient, easier, more straightforward, but if you don't train somebody how to use a system, um, they're, they're going to go back. To the old ways because they, they need to do their work. Um, and then the other one, yes, is there is a large component on the side of the, the technology that you have available and many of these organizations have some plants that have older technology that you need to bring up to a certain level, uh, versus, you know, a brand new plant that that was just built that has all of the latest bells and whistles, uh, and. You can get started uh with a predictive maintenance system, uh, if you have systems in place like data historians that are capturing. That real real-time data that that our system, you know, needs, needs to leverage. So companies that don't even have a system, you know, or something similar in, in place, uh, have more work to do to get ready um to take advantage of, of a platform like, like Sensei, um, but with that being said, Uh, there's still ways, uh, to get started, so, so don't ever not get started. Yeah. Yeah, no, no, no, that's great answers. So would you say, you know, even some companies who obviously aren't, aren't in the most. Most kind of prepared situation in terms of data, so they're not having all the data in like a organized system, it it doesn't mean that they can't reach out, that the support from the, the Siemens team obviously working on Sensite, predictive maintenance is still there, so it's, you know, it's not a deterrent factor, they can still reach out and we can still help them with that and make sure that, you know, all their um all their data and things can be prepped before the initial deployment. Yeah, that it, we have a multitude of, of avenues that that we can take to, you know, get the data, get the data in, in the right format. Uh, we just have to have those great conversations with our customers about meeting them where they're at, uh, so doing an analysis of, of what their systems, what their equipment is capable of, of today, and let's map out a path forward to get us where, where we want to go. Um, maybe there's some sensors that that we need to add. Maybe you do need to go out and get one of those other types of systems. Uh, we also have Siemens Insight Hub, uh, you know, that can be used, uh, in, in, uh, concert with, with Sensii. So there's, there's many avenues that that we can take and, and just, uh, get a company prepared to, to move forward. You also take a look at the, the different plants, um, and. Deciding where's a good place to start. Do we want to start with an older facility and help it get up to speed, uh, with, you know, some of the newer plants, or do we start with, you know, a plant that has some of the technology to prove out some of the concepts of predictive maintenance? So those are discussions that that we have with our, our customers, um, to help, you know, get a system like this, um, successful, uh, to, you know, on, on the first pass. Hm. No, sounds great. Thank you, Candy that that's that's really helpful. Um Yes, so I think that's pretty much everything on my side. Um thank you so much for those answers were super detailed. I did have a couple more questions, but you actually answered those within your other answers. So I I like I don't want to take up so much more of your time, so I'll stop recording now and then we have a little chat. I know I ended up shoving probably multiple things. So if you want to ask a question and get more of the 32nd clip, I mean, I I think we scheduled a half an hour, so I, you know, up to you. Sure. OK, well I've got to think on the spot now, so I'll I'll put you on the spot for 15 minutes is my g, but um, yeah, one, this might be, again, please don't feel like you need to have anything prepared, you know, we can edit things out, but um, how do you feel kind of the the sales team around the products like predictive maintenance is different to other predictive maintenance products like the service you provide, how do you think you differ from other, Other teams and other um other products. I, I think first the approach of trying to meet the customer where they're at and being able to say let's take advantage of of the data that you already do have and not come in with the assumption that you need a whole bunch of other systems to get started. With our system you have the ability to start getting quality predictions with oftentimes information that's already available inside of the PLCs and control systems, and we don't necessarily have to add a whole lot of hardware in order to do so. The second portion of it is is the fact that our models are embedded inside of our environment. And we don't require that our customers go out and spend months, um, or longer on developing the, the models, uh, themselves. So most of our customers, if not all of them, none of them have had to hire more staff. Or train their staff in different ways in in order to to use um our our software it's very geared towards the the maintenance team and making them, you know, effective in how they perform their their job functions. Please like and subscribe on all major podcast channels. And to find out more about Sensei predictive maintenance, go to Siemens.com.

Full transcript (WebVTT)