That’s the average loss an energy manufacturer experiences when its average-sized gas turbine stop working for a single day. That quickly translates to an average annual loss of 5 million dollars. In the automotive industry, halted production costs up to $250,000 for every 5 minutes. The average cost for 5 lost minutes is $110,000.
Whatever industry you’re in, it’s frustrating. You see communications, processing, cloud computing and artificial intelligence making a big positive impact on other companies’ bottom line, but your operations are still dependent on 1970s technologies, costing you an average of 17 days of machine downtime a year.
Moreover, modern manufacturers are obliged to strict production timelines and flawless product quality. In lack of better solutions, vendors and service providers spend billions of dollars on backup systems and redundancies just to mitigate the very precious production downtime.
In energy power plants, for example, billions of dollars are spent on asset reliability and triple redundancy.
But there are better – and much more cost effective – solutions.
Why the Traditional Factory’s Reactive Maintenance Will Keep Costing You Billions of Dollars
Industrial manufacturers are doing their best to monitor manufacturing floor production and asset performance and reduce costs. Yet there’s only so much you can do when your monitoring systems only alert you on failure-related events well after they happen, which is the case in traditional factories.
And that’s not all.
Traditional Monitoring Systems Discard Precious Data that Could Help You Take Action Before Machines Break Down
In the traditional factory, only a handful of key performance indicators were manually picked to be monitored. Then, a rule-based set of rigid threshold alerts was pre-calibrated by process control engineers.
There’s a wealth of data that’s generated between these pre-fixed thresholds – data that could help you see what happens with your machines at all times, notice when things are starting to go south, and do something about it before you lose machine uptime.
But this wealth of data gets ignored and discarded.
Preventive Maintenance Replaces Machinery that Might Not Need Replacing
Considering the broad array of assets utilities are responsible for — turbines, transformers, pumps, vehicles, miles of pipes and cables — and the risks associated with asset failure, a reliable maintenance strategy is a priority for them.
In a quest to ensure asset reliability, utilities have progressed from primarily a reactive, break and fix “repair” approach, toward a preventive maintenance approach. Now, machine parts get replaced as they age or on a schedule recommended by the manufacturer. This approach is facilitated by asset-tracking capabilities in ERP (enterprise resource planning) systems and by maintenance systems developed specifically for this purpose.
But the dire state of affairs is further augmented by the fact that preventive maintenance measures dictate the replacement of machine parts regardless of their actual condition, thus resulting in additional downtime and further maintenance expenditure.
How the Smart Factory Gives You a Competitive Advantage and Improves Your Bottom Line
Nowadays, even though most factories integrate computerized elements in their processes, they still struggle with the high complexity of planning and controlling the production process. Most processes are serial, hardware-oriented, and not sufficiently generic.
Here’s What You Need to Survive in Today’s Global Market
Increasing your profits and protecting your reputation in today’s industrial environment means stepping up your game.
You need to be able to provide high-end products and services at a high quality, minimal cost and in the shortest possible time.
As the global competition intensifies, requirements become stricter: products must be more personal, more market-oriented, and produced in even shorter life cycles.
Therefore, factories are required to operate in a leaner and more efficient way.
Smart Factory: Industrial IoT enables machines and components connectivity
How the Smart Factory Interlinks Machines and Networks to Give You a Competitive Advantage
Thankfully, new technologies – the Internet, the cloud, high performance computing, big data analytics and distributed systems – are making all of this possible, as the smart factory (sometimes called Industry 4.0) emerges.
The basic principle of a smart factory is that by connecting machines, components and systems, we create intelligent, interlinked networks along the entire supply chain, which can control each other autonomously.
Want examples for visions of a smart factory?
1. Self-organizing communities of machines that respond to unexpected changes in the production process (self-awareness).
2. Supply chains that automatically coordinate with one another.
Unfinished products that will send the required data for their processing to the machines which, in turn, will turn them into finished products.
3. Machines that predict failures and trigger maintenance processes (self-health-prediction) and result in near-zero downtime.
How the Previously-Discarded Data Your Machines Collect Helps You Move into Predictive Maintenance and Increase Machine Up-time
Remember that data your monitoring system discarded in the traditional factory?
The smart factory thrives on it.
In the smart factory, you use advanced information and communication technologies – such as sensor networks, cyber-physical systems, cloud computing, and big industrial IoT data mining – to generate, store and analyze all your manufacturing data. All this data gets processed with machine learning, deep learning, advanced algorithms and analytical tools.
That’s how you get valuable information about different aspects of factory work and the current status of each machine. Then, you use a knowledge discovery algorithm to detect and address unknown issues, such as machine degradation, in the factory floor.
From Reactive to Proactive: Repair Only Machines that Need it, When They Need it, and Before They Break Down
The algorithms in a smart factory obtain a combination of health information from various sensors. They determine the condition of in-service equipment, and predict when maintenance work needs to be performed. Alerts about required maintenance tasks are triggered at optimal times, promising significant cost saving.
In other words, transitioning from the traditional factory to the smart factory means no more preventive maintenance, which often requires significant, unnecessary expenses. No more replacing perfectly good parts when it’s not actually required, just because it’s on the schedule.
Transitioning to a smart factory means letting intelligence systems with a deep understanding of every machine in your fleet, and the relationship between them, make predictions and prescribe courses of action.
We call it preventive and prescriptive maintenance with built-in intelligence.
A benefit could be as simple as understanding what components are deployed in the field. When a mechanic is assigned to repair a component, he knows which parts and tools to bring to the job, which drastically decreases repair time and increases effectiveness.
But it’s usually much better than that.
Industry 4.0: The current industrial revolution
Let Your Machines Tell You What They Need
With predictive analytics and failure prediction, the smart factory lets you get ahead of equipment failures, which helps you avoid major breakdowns or outages.
Let’s say a pump component starts vibrating more than normal, or the oil pressure in a machine changes, indicating something unusual is starting to occur.
The built-in machine sensors pick up on that change and start a complex correlation and prediction process that eventually may emit an alert to the dashboard — or to a smartphone or another device — warning that the asset is projected to fail in the following few hours or days. This gives the dispatcher or operator a chance to schedule a repair to prevent costly downtime of the equipment.
By employing this predictive maintenance approach, utilities and manufacturers can maximize their equipment uptime. As a result, maintenance is performed according to what the machines tell you, not the calendar or a predetermined metric. If you go further and employ prescriptive maintenance, you can even get action recommendations based on what worked best in the past.
We can now repair or replace only the components that need it, when they need it, the way that works best for them.