Early Learnings from a Logistics Testbed
architecture . 11 Each tier is home to its own generic service type , which can be fragmented into several sub-types .
Services are currently the topic of many discussions in research and practice , for example against the backdrop of service-oriented architecture , service dominant logic and micro services . Services “ means applying specialized competences ( knowledge and skills ) through deeds , processes , and performances for the benefit of another actor or the actor itself . 12 ” This means that services are small , independent , and loosely coupled functions . In the context of data ecosystems , business models are realized through an interplay of services from different partners and these services are independent of the specific application and can therefore be integrated into very different ecosystems .
Within the context of the testbed Data Services , Information Services and Analytic Services could be identified . In the following text , six generic service sub-types are defined based on the experiences from the testbed .
4.1 THE EDGE TIER
The Edge Tier processes raw sensor measurements to provide a clean set of reliable data to the Platform Tier . We observed two different kinds of Data Services that facilitate this processing . A Data Service filters the positional data acquired through the gateways GPS system to exclude any data that does not meet predefined quality criteria . This quality filter removes data points that have accuracy below a certain threshold , no GPS fix , or where the ignition of the forklift has been off for the entire measurement timeframe . By doing so , the Data Service – Filtration only passes high quality data on and reduces the amount of data that gets transmitted to the Platform Tier .
Another Data Service creates a new data point from existing measurement data . It transforms the data to create additional data streams . We observed that the built-in accelerometer in the gateway was used in a Data Service to calculate whether the forklift is on-duty or off-duty . This type of local processing is what we call a virtual sensor as it requires the mutation and reinterpretation of existing sensor data . The Data Service – Virtual Sensor has therefore been used to approximate what could be a very complex sensor setup that directly measures if the forklift is currently in use or not . Virtual sensors do come with tradeoffs but provide a convenient way to enlarge the available data on the Platform Tier .
Data Services on the Edge Tier aim to clean and transform data . They make sure that only valid and useful data is transmitted to the Platform Tier . Data Services can expand existing measurement data by creating virtual sensors . Data Services can be physically located on the sensing device or on an additional Gateway that acts as a local edge hub .
11
Lin , Shi-Wan , et al ., 2022
12
Lusch & Nambisan , 2015 Journal of Innovation 19