Select Hardware for an IoT Solution in South Africa
Written by Morne Maree
In the pursuit of crafting an optimal IoT solution tailored to an organization’s unique requirements, the focus shifts to the crucial hardware component, marking the second phase of the IoT Full Stack.
If you missed the previous article on Hardware fundamentals for the IoT region, read it here.
However, this endeavour, as we’ll discover, isn’t as straightforward as it may seem. Challenges come into play during the hardware selection and procurement process, and here we will delve deeper into the intricate realm of IoT hardware, exploring both theoretical hardware elements and the practical considerations inherent in hardware selection within this specific geographical context. Unpacking the diverse array of IoT hardware elements, from IP-enabled devices to sensors and gateways, we’ll unveil the pivotal role each plays in a comprehensive IoT solution. Recognizing that the choice of hardware is intertwined with the broader IoT platform decision, we’ll underscore the significance of an informed selection to unlock data’s inherent value and enable seamless integration. Stay tuned for a comprehensive guide on navigating the intricacies of IoT hardware selection, ensuring your organisation’s IoT journey is poised for success.
Non-Technical considerations
Vendor/parent company
You need to ensure that the Device you are choosing comes from a reliable and credible vendor—one that has a record of accomplishment in your market. It is always a good thing to use a vendor with local representation and support. Trust is a key factor in the IoT device market—you are trusting a device to generate your data after all. It is important that you receive support in the same time zone you are in. In the South African, SADC region case, will you get local in-country support? It is quite easy to buy and implement a device from an international vendor online these days, and in the short term the suitability of the device could look compelling. It is important to ask how the device selection may play out in the long term.
Technical
Greenfield vs brownfield IoT solutions
Contextualising this to the industrial IoT, greenfield IoT projects involve deploying plant or building technology simultaneously with new IoT Hardware Devices, usually by the original equipment manufacturer or a distributor, whereas brownfield IoT projects deploy of IoT Hardware Devices on existing plants or buildings that are already in operation, by a Full Stack IoT vendor or systems integrator like Skills Revolution Digital. In South Africa and SADC region most of the IoT Solutions needed and implemented are Brown Field Projects.
Consideration: So, if you are implementing a brown field project, is it technically possible to “rip and replace” the legacy equipment and is it financially viable?
Sensor
Understand very clearly the parameter or set of parameters that you need to measure or monitor. There are several devices capable of achieving the same goals and extracting the same data. It is critical that the correct device is chosen that is most appropriate for the use case.
Considerations: Accuracy, repeatability and speed of measurement is critical. The device sensor must be ‘fit for purpose’ as there is always a consideration for cost.
Data acquisition and send frequency
In many IoT discussion the term “real-time” is used, however, Data acquisition (DAQ) gathers analogue information at a fixed time interval (the data sample rate) and transmits it as a digital signal to a remote output device for a digital readout. DAQ may include signal conditioning (to manipulate and scale raw sensor readings), and analogue-to-digital converters to convert the analogue sensor readings into digital values so that they can be processed and analysed.
Considerations: Understanding that the value measured and visualised is never real time. Understand how fast the parameter that needs to be measured, changes, and choose the measurement rate accordingly i.e., Temperature is normally a slow-moving variable, so it is not necessary to do hundreds of measurements per second. The frequency of data collection has costs associated with communication, data storage and analytics.
Mobile or Fixed – Wireless or Wired communication, battery or mains powered?
Will the parameter that is being measured or monitored be moving (mobile application like the temperature of a cold storage truck) or is it going to be fixed at one place (energy measurement of a building).
Considerations: Connection stability, Security (Susceptibility to hacking), Accuracy, Cost, Mobility, Aesthetics, Installation Flexibility and Cost, Existing Networks, RF Interference, Signal Blockage or Congestion…
Edge or Cloud Computing
Cloud and edge computing are different, that cannot replace each other. Edge computing is used to process time-sensitive data, and cloud computing is used to process non-time-triggered data. Cloud computing is centralised, with servers stored in faraway, large-scale data centres.
Considerations: Latency, Time-sensitivity, Mission Criticality, Required Processing Power, Connectivity, Bandwidth, Cost
Availability of Onboard Storage
Data storage: This layer stores data collected from sensors and devices at the edge or cloud for long-term or short-term applications. The edge gateway provides functionalities, such as sensor data aggregation, pre-processing of the data, and securing connectivity to the cloud.
Considerations: Network Stability, Data Redundancy, Data Recovery; Does the data collected contain cumulative information?
Environmental and Aesthetics (Indoor or Outdoor Use) – Enclosure design
IoT Devices not only have to be Safe and secure, along with the weatherproof factor against the possible elements but must also be aesthetically pleasing.
Considerations: Indoors or outdoors, protected from heat, dust, water, moisture, and accidents, Intrinsically Safe, Installation method,
Depending on different use cases, the intent of safety needed also varies, and it is a crucial manufacturing decision for every IoT device company.
Device management
An IoT device management platform is an integrated application that simplifies IoT device management by allowing you to manage and monitor the entire lifecycle of devices and sensors in one place—from planning and onboarding, to monitoring and maintenance, through to retirement.
Considerations: Can you onboard the device, is configuration and control possible, can the user monitor and diagnose, perform maintenance, decommission, or remotely update?
Conclusion
In conclusion, the journey towards designing effective IoT solutions for the South African context entails a meticulous exploration of hardware selection, which marks a pivotal phase in the IoT Full Stack. The challenges and intricacies of this phase cannot be underestimated, as they encompass a range of non-technical and technical considerations that directly impact the success of the IoT implementation.
Within the realm of non-technical factors, the trustworthiness and reliability of hardware vendors play a critical role, particularly in the context of the SADC region. Local representation and support become key determinants of long-term viability. The significance of these factors cannot be overstated, as they dictate whether a seemingly advantageous device choice will indeed stand the test of time.
On the technical front, the distinctions between greenfield and brownfield IoT solutions emphasize the importance of assessing the feasibility of integrating new hardware within existing operations. The choice of sensors holds paramount importance, as their accuracy, repeatability, and compatibility with the specific use case directly impact the quality of data collected. The frequency of data acquisition and transmission must be finely tuned to balance real-time needs, communication costs, and data storage capabilities.
Factors like communication methods, power sources, mobility, and installation flexibility bring into focus the intricate balance between wireless and wired solutions, as well as the choice between edge and cloud computing. Each decision impacts latency, processing power, connectivity, and cost.
The storage of data, whether onboard or in the cloud, is not just a technical matter but a strategic one, influencing data redundancy, recovery, and cumulative insights. Moreover, the environment in which IoT devices operate has a substantial bearing on their design, encompassing considerations of safety, aesthetics, and protection against the elements.
Lastly, effective device management emerges as a critical component of the IoT hardware landscape. A robust device management platform can streamline the entire lifecycle, from planning and onboarding to monitoring and retirement.
In the South African and SADC context, where unique challenges and opportunities shape the IoT landscape, navigating the hardware selection process requires a comprehensive approach that bridges non-technical and technical considerations. Armed with this comprehensive understanding, organizations can embark on their IoT journey with the confidence that their hardware choices are well-informed and poised to unlock the true potential of data-driven innovation.
To delve even deeper and access a wealth of insights, we invite you to explore our website: Skills Revolution Digital. Or stay tuned as we explore each layer of the Full IoT Stack and navigate the complexities together, to ensure that your IoT venture is poised for triumph.
Olifantsfontein, Gauteng, South Africa. | +27 76 693 9010 | digital@skillsrevolution.co.za