Advanced Sensing and Blockchain Technologies Optimize Supply Chain Readiness
As the COVID-19 pandemic disrupted supply chains in vital industries, the CTMA Program designed an initiative to prevent problems with the production and distribution of critical supplies in the event of future pandemics, natural disasters, cyberattacks, and emergencies such as chemical, biological, radiological, nuclear, and high-yield explosives (CBRNE) situations.
The collaboration—Advanced Sensing and Blockchain Technologies to Optimize Supply Chain Readiness—brings together two industry partners, Presage Technologies and SIMBA Chain, along with the Assistant Secretary of Defense for Acquisition Enablers (AE) and the Deputy Assistant Secretary of Defense for Materiel Readiness.
Since the initiative’s launch in September 2021, the team has been using the COVID-19 pandemic as a case study. They have performed two mutually reinforcing lines of work to demonstrate how the latest advances in data science can generate vital information. First, the team is improving the collection of population health signal data by documenting changes in the health, mobility, and economic activity of given regions that impact suppliers. This effort leverages advanced sensing and aggregating technologies, along with artificial intelligence and machine learning (AI/ML) algorithms to help analysts predict future disruptions in the supply chain. Second, the team is integrating this population health signal data into commercial blockchain technologies. As a result, they can help to automate supply and demand logistics decisions, enabling just-in-time inventory management of critical products and equipment.
“Any public health information system that records COVID cases takes about five days for the process to even reach the system, but this technology provides the information immediately,” said Ian Taylor, CTO and co-founder of SIMBA Chain, which developed the project’s blockchain tools.
“We think of this technology like a weather service,” said Mark Oliver, President and CEO of Presage Technologies, which builds software-based medical devices. “We were calling it ‘health weather’ for a while.”
To provide the “health weather,” Presage Technologies’ team of biomedical and software engineers use remote photoplethysmography (rPPG), a camera-based, unobtrusive technology that allows continuous monitoring of changes in vital signs to help diagnose and treat diseases earlier. This technology can be used to gather anonymous vital sign metrics—pulse rate, heart rate variability, breathing rate frequency and volume, and blood oxygen saturation—from cell phone cameras.
“Imagine pulse oximeters or respiratory frequency monitors. We can do all of that now with just the camera on a phone,” said Oliver. “We’re achieving clinical-grade accuracy and we produce vital signs metrics from videos in a privacy-oriented way. We aggregate videos into huge, square-mile blocks so that no one can ever point the health data back to any individual.”
By collecting data exclusively from open and social media sources, the team solved a significant problem in assessing global health risks: gathering and modeling population-level biometric data in a manner that does not rely on host nations’ capability or willingness to share information, all while maintaining robust individual privacy protections.
“We don’t want to turn this into a Big Brother tool that says whether someone has COVID-19 and possibly make them take a temperature test or kick them out a venue,” said Dan Janes, Government Engineering Lead at Simba Chain. “We really wanted to approach it from a public health perspective by focusing on the aggregated data for a particular location.”
The team applied algorithms that provide epidemiological projections of infection rates in a specific area. By using a deep learning model, the team forecast the actual COVID-19 infections per 100,000 people, using Germany as a case study. The collaborators checked their model against the actual infections per 100,000 people in Germany, which is reported every seven days. The results were stellar: 80% accuracy with a very low error rate, without requiring any on-the-ground information.
The project uses a commercial off-the-shelf application programming interface (API), a mechanism that enables two software components to communicate with each other. For example, the weather bureau’s software system contains daily weather data. Weather apps on cell phones communicate with this system via APIs, then display that data on the apps.
“Every day on the API we make available the ‘COVID weather,’ or the number of infections per 100,000 for the whole country,” explained Oliver. “Any entity that we give access to our data can ping our API. Then, they can pull that data into the blockchain. From there, they have the demand-sensing piece. Now they can look at inventory and make sure, say, in the case of the flu, there is enough Tamiflu on hand.”
This data facilitated the project team’s second task: using blockchain to automate the ordering and distribution of PPE and other health protection logistic items to areas in need of them. Blockchain is a digital, decentralized, and publicly accessible database that records transactions (blocks), across many computers, that are linked using cryptography. Each transaction contains a cryptographic hash of the previous block, a timestamp, and transaction data. As a result, data cannot be altered retroactively.
While blockchain is widely known for its role in crypto-currencies, it is being used in multiple industries and is especially useful in supply chain management because it provides immediate, transparent access to information stored on an immutable ledger that can be accessed only by permissioned network members.
How does this technology work? First, SIMBA Chain pulls data from Presage’s API once a day. Next, the data is fed into the SIMBA Extract, Transform, Load (ETL) Pipeline—a process that packages the data and pushes it to SIMBA’s Blocks platform. Then, Blocks sends the data to a SIMBA-created smart contract, which notifies the end-user of any alarming COVID-19 predictions that pass thresholds set by the end-user.
“The way blockchain works is that a transaction comes through a smart contract, which is a piece of code, that you can add logic to, and that results in something being recorded on the blockchain, which is not changeable,” said Taylor. “The data coming in and decisions being made on that data are being made available to the supply chain in a way that is completely transparent, trustworthy, and un-hackable.”
Blockchain facilitates automated inventory management because it can be connected directly to an ERP system to automate the ordering of needed supplies.
“If the COVID-19 rates are ramping up quite a lot, the system might order masks and other supplies, then have them shipped to that geographical region automatically to preempt the situation that’s about to occur,” said Taylor. This will ensure that PPE, medical supplies, and other vital goods are available where they are most needed.
While this project produced a deep learning model to forecast COVID-19 infections, the model can be used to forecast any infectious disease. For the DOD, this population health signal data will help leadership make judgments about population and community health risks to inform force-health protection decisions worldwide. This data will also help to better inform procurement and supply chain decisions so that correct parts, medical supplies, PPE, and other items are available to protect our armed forces.
Additionally, this technology could help government leaders make informed decisions to ensure that hospitals and other critical infrastructure remains operational; to sustain the unimpeded flow of food, water, medical supplies; and to safeguard basic quality of life products and services.
Moving forward, the team plans to enhance the technology to support more locations, create data visualization, provide stronger supply chain integration, and build a user interface (UI) to help end-users interact with the system.