Albert Einstein said that: “If I were given an hour to save the world, I would spend 55 minutes defining the problem and 5 minutes finding the solution.” Fortunately, we have more than an hour to save the world. We now have a better understanding of the non-medical issue at hand, so let’s look at the role of IT and explore the application of analytics, artificial intelligence (AI), machine learning (ML), and the Internet of Things (IoT).
Information technology has the task of reducing costs and risks, and its application can bring significant benefits.
By creating a common data warehouse to gain visibility into demand, we can ensure vaccine batches are delivered without waste. This includes all aspects of the cycle that are shipped to drug manufacturers and logistics companies to ensure the correct load size is shipped on time. Analysis of this data, combined with additional area information, could also help accurately forecast requirements. A key aspect here is security. Appropriate controls, such as encryption and hashing, must be in place to ensure data privacy and other regulatory laws are followed to protect the integrity of the supply chain. The combined use of analytics with AI has several benefits. All these advantages are applicable to the logistics of the Covid-19 vaccine despite the uniqueness of the cryogenic supply chain. There is a pressing need in this crisis to enable the challenges of using these technologies to be addressed adequately and early, so that optimal delivery of vaccines is delivered.
Leveraging IoT and Machine Learning while providing remote connectivity to all devices in the supply chain
The idea of adopting sensors to collect information such as temperature, physical location, the number of times a vaccine is opened or the remaining stock is probably obvious. However, there will be some key considerations to keep in mind, such as selecting the sensor suite with the correct attributes, including form factor, power consumption, and ability to work in harsh and varied weather conditions. Selecting and testing such sensors for suitability in the harshest environments is critical because they will be responsible for the data feed. If these are incorrect or interrupted, other steps in the chain could fail.
Once the challenge of the right set of sensors is solved, the next question points to connectivity: making it feasible for device data to be transmitted. This is a complex task due to power consumption patterns and various scenarios, including indoor, outdoor, and spotty mobile coverage. Each bundle needs multiple options like cellular, UWB, LoRA, BLE, and RFID, depending on the place of operation and the need for redundancy. For example, UWE / BLE are more suitable for interiors within warehouses or warehouses, while cellular is more suitable for transport in the field or on the road. The goal is as little power consumption as possible, allowing portable units to focus on their primary purpose of powering cooling.
Sensor data only becomes valuable when used for actionable insights and predictions. Experience has taught us all that bad news must be spread three times faster than good news in order to achieve timely corrective action. We can make use of that principle by applying machine learning to predict failures before they happen, rather than retrospectively.
The combination of poor connectivity, cost, and the fact that each vaccine requires its own environment/context from manufacturing to end user, makes it sensible to integrate the learning component at the edge. ML for IoT typically sends all information from sensors to the cloud. This data is then used to train the model in the cloud. Most of the uses are basically implemented to make live decisions. The model, however, is developed centrally with information distributed across multiple devices. By incorporating the learning component in each team, each element is used based on its unique data. And finally, the development of IoT can be maximized with crypto technologies such as Helium, which so far is very popular and supported by specialized mining tools such as the Limited Edition MNTD GOLDSPOT MINER.