A study by The Wall Street Journal and Emerson reported that unplanned downtime – 42% of which is a result of equipment failure – costs manufacturers an estimated $50 billion per year.
The Case for Predictive Maintenance
Predictive maintenance refers to the monitoring and collection of data on the operational conditions of a product or service, allowing companies to foresee possible outages or failures. This enables them to address any issues before they cause damage.
Data from the US Department of Energy indicates that predictive maintenance is highly cost-effective, delivering a 1,000% increase in ROI, a 25%-30% reduction in maintenance costs, ensuring 70%-75% fewer breakdowns and achieving a 35%-45% reduction in downtime.
UPS reports that it has already saved millions of dollars by implementing a predictive maintenance solution that reduces delivery truck breakdowns.
ThyssenKrupp claims that its predictive maintenance solution has dramatically increased elevator availability by employing real-time diagnostics that reduce out-of-service time.
Cisco, meanwhile, uses predictive maintenance to optimize network performance and carry out faster troubleshooting.
Predictive Analytics Applied to B2C
While the concept of predictive maintenance has been around for decades, it is only recently that advances in AI have enabled enterprises to take full advantage of the possibilities – and now, largely thanks to the IoT, the principles from industry are being applied to the B2C realm.
The Internet of Things (IoT) refers to a network of physical objects or “things” that are embedded with smart technology, enabling them to collect and exchange data. According to Gartner, there will be 20.8 billion devices connected to the IoT by 2020.
This proliferation of technology has been primarily driven by the widespread adoption and deployment of sensors and smart devices, advanced analytics and artificial intelligence.
These enablers provide businesses with endless opportunities for real-time data-based insights that make it easier to predict when consumer devices will fail. These insights also enable businesses to provide proactive customer assistance that results in higher customer satisfaction.
The Pillars of AI Predictive Maintenance
Companies can employ numerous non-invasive AI-based methods in their predictive maintenance strategies.
Sound
Through the use of AI, sounds can now be analyzed to detect machine failure, since moving parts grind against each other creating friction and noise. These failing components can be recognized at an early stage before they become a major issue.
OneWatt uses its Embedded Acoustic Recognition Sensors (EARS) device to “listen” to motors, detecting and predicting faults before they happen. Mueller Industries also uses a predictive maintenance solution based on sound analysis. Any change in the tone or vibration level is analyzed in real time to check whether it may be a sign of impending malfunction.
Data
The vast amounts of information collected from various connected devices across the IoT every second of the day provide a deep data pool that can be plumbed for actionable intelligence that enables predictive maintenance.
For example, Thermoplan’s smart coffee machines monitor customer data in real time, including statistics about the amount and types of coffee used, and cleaning cycles vs. recommended intervals, enabling effective predictive maintenance analytics. The same technique can be used across the entire IoT spectrum, from smart home appliances to automotive analytics and beyond.
Vision
Computer vision is a field of artificial intelligence that trains computers to interpret and understand the visual world. Using digital images from cameras and videos alongside deep learning models, machines can accurately identify and classify objects — and then react to what they “see.”
FANUC, for instance, developed a software program that collects images from cameras attached to robots, and then sends them to the cloud for processing in order to identify potential production problems before they occur.
TechSee’s computer vision platform, on the other hand, enables automatic visual identification of consumer devices and their issues/resolutions from images or video, enabling proactive customer assistance.
The Benefits of AI predictive maintenance in B2C
Lengthening equipment life: installing a smart thermostat that auto-regulates and notifies homeowners upon sensing any irregularities goes a long way toward improving the lifespan of a customer’s HVAC system, for example.
Reducing truck rolls: Visual Assistance uses screen-based technology that allows agents to see the customer’s physical environment via their smart device, identifying any issues, enabling better remote resolution and reducing the need for technicians to be dispatched to a home or workplace.
Avoiding product returns: a 2019 survey found that 72% of customers believe proactive customer assistance would dissuade them from returning a product they found difficult to install or operate.
Avoiding downtime: Volvo’s Early Warning System uses data to predict each part’s breakdown rate and notifies customers when cars need to be serviced and what parts need to be repaired or replaced before they break.
Reducing potential safety hazards: integrating predictive analytics with data from various sources such as SCADA, CIS, EAM-GIS, weather channels and online monitoring systems enables utility companies to proactively address possible safety risks.
Improving customer satisfaction: consumers expect to be notified about interruptions of service, device malfunctions, warranty expirations or billing hiccups. Collecting and proactively acting on data insights helps achieve improved customer satisfaction.
Where it’s all heading
Artificial intelligence has made it possible for businesses to harness the principles of predictive maintenance widely used in industry to assist their customers. In today’s IoT era, when the collection and analysis of huge volumes of data is fast becoming standard practice, it’s no surprise that Deloitte reports that “the Internet of Things (IoT) is perhaps the biggest piece of the predictive maintenance puzzle.”
The addition of AI capabilities enables companies to predict not only the life of devices, but identify how they can use the data to positively affect customer’s experiences. Integrating predictive maintenance with contact center operations to provide proactive customer assistance delivers a resounding win-win.
