IOT and Servitization in Manufacturing-A Quick Guide to Key Technology Enablers
By PavanK.Malladi, CIO and Head Information and Communications Technology Sales, Ericsson India
When a Swedish manufacturer of bearings introduced 'sensors' in its bearings which would record machine rotations to provide a unique service to its customers on predicting machine life and productivity via this information, the entire world recorded these phenomena as 'Servitization'. Servitization today is used as a word to describe introduction of intelligent services on top of a product to extend the customer relationship that the seller of the product has. Not only does it introduce life cycle services as an additional revenue spinner, but also provides a mechanism to continue the customer relationship beyond sales and repairs it to another level. However, most if not all of the cases require the use of sensors to collect data, analytics to be applied on that data and a service to be created on top of that data to create such services.
The triple bottom line of people, profit and planet has never been more important than it is today. Some of the key benefits seen especially related to the manufacturing process:
*Better Signaling of Demand and Supply Chain Efficiency: IoT can enable better signaling of service level of the product at any time thereby increasing both the demand signals accuracy as well as supply chain efficiency by tracking shipped items for the purpose of returns, warranties and predictive support of failing items.
*Accelerated Innovation and Proactive Product Support: By providing continuous and real time data to manufacturers, IoT can enable continuous innovation via product usage tracking and behavior analytics.
*Sustainability and Safety: IoT-enabled sustainable operations can provide real-time visibility to energy and resource consumption and resource or material movement.
While the benefits can be expanded deeply with further imagination, the key to achieving this, the technology to enable IoT requires complex understanding of both Information (from an analytics view point), Communication (from connecting the devices to service innovation systems) and Sensor electronics (from building the sensors to work with the products).
Firstly, let's clear the air on what is the difference between Machine2Machine Communication and the IoT. The paradigm of sensors and analytics being applied across smaller devices interconnected across small areas by controlled networks is what we would know as Machine to Machine communication (M2M). However, if it's across a large set of devices interconnected across vast geographical areas; the default mechanism of communication chosen will almost need to communicate in a world-wide network: The Internet. When we contemplate following this path into the future, we find that nearly all everyday things will become smart nodes within a global network. This phenomenon is called the 'Internet of Things' (IoT); a trend that will almost certainly find its way into industrial production.
Let's now look at three things Sensor Electronics, Analytics and Communication technology, to figure out why IoT is now possible than a few years back.
Sensors- Electronics and Software leading
A distinguishing feature of the new technological environment is the transition to mechatronic systems. Electronics will be a fundamental component of future products, while hardware will be increasingly standardized. The major features determining the functionalities will be created by the software. In this way, traditional machine elements are transformed to mechatronic systems. For interconnection, the key mechanism in all the new objects will enable wireless connection, to allow mobility and will have its own unique IP Address in the IoT scheme of things. Such machines are now even being called 'Cyber-Physical-Systems'. One of the other distinguishing features is the 'context-sensitivity' of these machines.
Machines can now be extremely context sensitive due to their connection to the ambient environment surrounding them. Machines can respond via motion sensing around them thereby providing safety. This leads to more deeper mechanisms of control and efficiency often now referred to as Autonomous Optimization' used as a design principle in designing such systems.
IoT Analytics is about Small not Big Data:
Analytics and IoT concepts are much talked about, but not always understood or distinguished between in a stringent manner. Big data is a term describing large amounts of data, collected from a variety of sources, analyzed with the purpose of building business advantages. While IoT can add new types of sources to the mix, it does not just make up another incremental development step for big data. Just like the internet is more than a library of data to be used for analytic purposes, IoT is more than a new range of sources to add to the complexity. RFID provides identification, sensors provide sensory information, and embedded systems read things, digitizing all new sorts of information to be combined and analyzed.
Another important aspect of IoT that separates it from big data is the actuation purpose of many IoT applications. In IoT applications, the small data is more relevant than the big. The more important concept therefore is collecting small information, processing it, analyzing it in context and then immediately actualizing it that produces value in IoT. Thus, small, not big data communicated in real time by machines act autonomously by communicating between sensing and actuating equipment is key, and then recording the transactions for statistical quality control perspective or operations research is the need for IoT in Manufacturing.
Communication in IoT: Cellular with Capillary Networking is the solutionIf you analyze the needs for IoT, both in technical and business terms cellular technologies working in licensed spectrum are well suited to the application. Cellular offers advanced functionality when it comes to quality of service and security. 3GPP technology already has mechanisms to assign priority access to certain classes of device (such as smoke detectors) to make sure data gets through even in emergency situations when there may be network congestion. Security is an integral part of 3GPP cellular standards, and includes secure authentication and encryption of traffic. The key needs are satisfied as below:
Mobility: From fast moving vehicles, through slow moving humans (or delivery drones!), all the way to stationary garage door controllers that need no mobility at all. All these types of devices can be connected via Cellular Networks.
Bandwidth: Human-centric media devices such as TVs consume and generate a lot of data at great speed, while sensors might only transmit a few kilobytes per day, a cellular network can dynamically adjust the bandwidth requirement.
Latency: Cars communicating with each other on the highway need extremely low latency links, but also humans having a voice call, or playing a cloud-hosted game, cellular is the answer here.
Reliability: Connectivity for a medical sensor supervising a patient at home can literally be a matter of life and death, but less critical connection failures are increasingly frustrating and disrupting as well. Reliability is a key feature in a cellular network.
Availability: Similar, but subtly different from reliability: e.g. a sensor that is batching its transmissions of measurements may only need to be connected a few minutes per week, while surveillance systems require constant connections. Both may require a very reliable connection but don't have the same requirements on it being available all the time. A cellular network can be accessed at will without elaborate provisioning required for a more fixed network.
However, the only drawback with a cellular network is the high power requirement which is not necessarily a feature of the low power source devices.
This problem is therefore solved by 'Capillary Networks' which essentially is a phrase drawn out of biology, since it is most commonly used to describe the smallest and thinnest parts of the network of blood vessels in animals. The term here is used in IoT to describe complementing radio technologies, often short-range and low-power, which connect to the macro cellular network through more traditional technologies.
In effect, there are several access networks: wireless sensor networks to measure the physical world, the cellular networks (including 3G/4G) to handle mobility, mesh networks to support new applications and services. When we aggregate all these networks, we form a capillary network. Therefore, capillary networks are a key to building an effective communication network to clearly complete the IoT cycle.