1.3 Domains and technology – BDAFs in context
Another useful way to categorise BDAFs is through domain areas covered and the technologies that are used in big data activities. For example, BDAFs can be categorised in the following ways (Mohamed et al., 2020):
- Sources of data generation (for drones, this is the ‘Internet of Things’ or IoT, but sources also include the world wide web, phones, email and other digital communication, social media, sensors and mobile phones)
- Data format (unstructured, semi-structured and structured data)
- Data processing and storage (batch, stream and hybrid processing)
- Different types of data analytics (e.g. statistical analyses or visualisation techniques)
- Data visualisation (e.g. tabular, reports, dashboards).
Different domains will have different challenges and needs with respect to a framework. For example, in forestry management, data will often come from multiple sources. Drones provide just one of the sources of data, which can include other digital methods and physical analysis, such as chemical and genetic analysis (Taylor et al., 2020). This thus influences how the analytical approaches identified in the BDAF will account for different types of data. For example, researchers at the Norwegian Institute for Bioeconomy Research (Puliti and Granhus, 2021) compared the usefulness and cost effectiveness of drone technology in monitoring forest health with other, more established methods, such as piloted airborne laser scanning (ALS) and traditional, field-based methods. The researchers found that using the right analytical tools was an important element of transforming raw data collected by drones into ‘actionable insights’. As actionable insights are what help businesses and their customers to extract value from big data, both the quality of the data and the quality of the analytical tools will be crucial aspects to include within a BDAF.
As with the drone delivery case in Activity 1, the security of data will also be a key consideration for businesses and their customers. Data security is also prioritised by government and the public sector. For example, above and beyond solving technical issues related to using drones to deliver small parcels to remote rural areas, the personal customer data utilised in requesting, paying for and making the delivery will need to be secured, including fleet security. If these data need to be shared between different businesses (such as between a retailer/shop owner and a drone delivery business), how will this be managed?
There are, of course, more common delivery models already in place that can be used as data sharing models, but how these are implemented within a scenario using drones needs careful consideration. Researchers note that drones can be considered part of the Internet of Things (IoT) and are thus themselves at risk of internal and external threats: ‘As part of heterogeneous networks, things have to support advanced security concepts, such as authentication, access control, data protection, confidentiality, cyber-attack prevention, and a high level of authorization’ (Lagkas et al., 2018, p. 2). How a drone business ensures data security is essential in establishing trust with their customers and with regulators.
Activity 2
An essential outcome of the ICAERUS project is the creation of an open access platform that includes a Drone Data Analytics Library, where algorithms and data sets from the Use Cases and other drone innovators can be shared with the global drone community.
Go to the ICAERUS Drone Data Analytics Library [Tip: hold Ctrl and click a link to open it in a new tab. (Hide tip)] on the ICAERUS website and have a look through the available data sets and algorithm pages.
a.Consider how the uploaded data sets relate more generally to the desired outcomes of the different Use Cases. You can read more about the Use Cases on the ICAERUS website. How are the data and/or algorithms featured in the expected outcomes of the Use Cases?
- b.Using the project’s livestock monitoring Use Case in France, consider the value data provide to farmers.
- c.How might the different Use Cases (and their data sets and/or algorithms) relate to your own drone business idea?
Discussion
The livestock monitoring Use Case in France is assessing ‘labour reduction capabilities of drone-based herd monitoring’ and is examining ‘governance models and drone adoption barriers and drivers’. The Use Case aims to achieve the following outcomes:
- Quantify the cost, labour and time saving impact of drones for livestock monitoring.
- Identify drivers and barriers for drone adoption.
- Support grazing livestock systems to better meet society’s demand for free-range animal products.
- Increase farmers’ quality of life by improving work conditions.
Each of these outcomes can be thought of in terms of extracting value from big data, i.e. the data sets, algorithms and drone technologies. Successful outcomes, such as quantifying the cost, labour and time-saving impact of using drones, will be influenced by the relevant ‘Vs of big data, as well as by the broader BDAF categories. For example:
- Volume: How much data will be generated and how will data be stored and communicated to the farmer? (Sources of data generation, data format)
- Veracity: Can farmers be assured that the data are giving them reliable information about the health and welfare of their herds? (Sources of data generation, data analytics)
- Variety: Are there different types of images or analytics that the farmer can use to make decisions? (Data visualisation and data analytics strategies)
- Velocity: Are the data processed and available in real time to the farmer to make it a time-saving alternative? (Data processing, sources of data generation)
As you learned earlier in this section, both data and analytical tools (e.g. algorithms) are important contributions to the value of your business idea. The ICAERUS Drone Data Analytics platform is a repository that shares data sets and algorithms from the project partners and open call grantees. You can download algorithms and practise with the data sets provided to see how they might be used in your own business case scenario.