1. Work performed (in accordance with the Work Packages referred to in the INNOSUP program)
Besides successful onboarding as well as internal and external training of the Innovation Associate (IA) the work in a first step focused on exploring the environment to understand the state of the art in physical tracking, screening, verification and assessment etc. as well as its limitations and identify the main players as well as the best practices and applications. This included the structuring of the screened input and a development and update of the research roadmap and in consideration of our guidelines and workflows, pilots, projects and services as well as customer requirements and requests.
In a second step, we proceeded with the development of business cases related to physical tracking and Artificial Intelligence / machine learning (AI). This in particular included to explore and develop relations to third party suppliers for physical tracking and AI tools and services. In addition this comprised internal documentation, communication and training as well as the preparation of management guidelines and decisions based on the findings.
In particular we developed an internal database and business matrix to support our consulting and project related activities. We considered the majority of tracking technologies to be part of the IoT ecosystem and analysed their key features, benefits and use cases in the mining industry as well as downstream the supply chain. For example, we have analysed the following most common physical tracking technologies:
- Marking technologies, from DNA to ink markers Tag identification
- Labels/tags attached to tracked asset: QR/Barcode Identification Labels
- Tag Identification using GPS/Bluetooth Low Energy/RFID/ NFC
- Asset Identification with 0G, LT 1 Low power wide area network
- Telco network services: Geolocalisation, Mobile coverage, 5G with IOT
- Hybrid positioning technologies blending: Wi-Fi Signals, Cellular Signals, GPS
- Satellite observation & drones
- Private radio solutions such as Motorola Solutions Tetra/DMR
- Mobile app providers
- Computer vision/AI and video surveillance
- Scan Technologies to transfer paper documents to digital data points
Besides building internal expert knowledge, our research also identified and assessed potential ecosystem partner companies and strategic alignments. Furthermore the communication to third parties, for example via publications as well as participation in online events has been planned and performed.
Moreover, to explore AI related use cases and to develop related business products and opportunities, our research focused on developing use cases, methodologies, test scenarios and processes as well as establishing a test environment to perform machine learning tests. We used AWS services, in particular the machine learning services and test environments to explore and implement the technology in a Minespider environment. This in particular included the AWS machine learning techniques and services such as Textract, Recognition, Comprehend and SageMaker.
2. Main results achieved
Work resulted not only in building an extensive internal knowledge base with the Minespider Team but also extended our network to experts, service providers and partners, strengthened our relationships to customers and supported us in the acquisition of new customers. This has contributed to creating an international ecosystem of various industry players that enhance the blockchain landscape. Moreover, we developed a taxonomy of physical tracking technologies in the context of the Internet of Things and a blockchain based supply chain tracking environment.
The main (non-confidential) findings were presented on the RawMat 2021 event and in an expert article published on the Material Proceedings (DOI 10.3390/materproc2021005001). Also we published non-confidential datasets to the suggested taxonomy of tracking technologies to effectively capture and input key data on the blockchain (DOI 10.5281/zenodo.5510478).
