The GoRFID Project is a strategic SAP project, which targets the development and evaluation of performance critical applications within the EPCglobal Network Architecture.
Visibility and real-time awareness are the two major use cases for the implementation of RFID in supply chain management. Objects, assigned with unique identifiers (Electronic Product Codes, EPCs) travel from production facilities to the consumers, producing data, highly relevant for most supply chain processes. Such data need to be stored and exchanged among different, potentially independent, supply chain parties.
The EPCglobal Network Architecture is a framework, which provides standards and specifications for information systems, targeting at the efficient processing, storage, and exchange of EPC information. Among others, it defines a component, called EPC Information Services (EPCIS). This software system provides a repository for storing EPC-related information, such as the EPC itself, temporal information, business data, and location data. Additionally, the EPCIS standard specification defines interfaces to capture and query the data stored in the repository. The EPCIS is expected to be deployed locally, within company borders, collecting only information from local data sources (e.g. RFID Readers). In modern globalized supply chains, products are handled by many different companies. Hence, the information about item travel is distributed over a number of local EPCISs. In order to foster company overlapping B2B collaboration, an additional component is necessary. Such component is the EPC Discovery Service (EPCDS). Given an EPC, the EPCDS resolves the addresses of all local EPCIS servers, which are in possession of information about the particular EPC. This functionality enables interested parties to track and trace objects even over company borders, creating full supply chain visibility. Unfortunately, at present, there is no standard specification for a Discovery Service component.
Looking at real world scenarios, the operation of RFID-enabled supply chains produces massive amounts of data. In our research, we investigated features and characteristics of the European pharmaceutical supply chain. Our investigations resulted in an analysis, which derived a total number of about 3,000 produced scan events per second and about 2,000 queries. Such numbers show that implementations for EPCIS and EPCDS need to be designed carefully, regarding capture and query performance.
Most of the existing implementations for EPCglobal components fall short regarding performance aspects. In this project, we leverage in-memory technology to design an EPC Discovery Service and an EPCIS Repository prototypically, focusing primarily on performance aspects. Using SAP’s in-memory technology HANA, we are not only able to improve pure transactional performance; we are also able to implement instant analytics on the stored data. Information sets, comprising of more than a billion records can be searched and aggregated within seconds and providing clients with completely new opportunities regarding the interaction with RFID-related data. For such analyses, we use SAP’s Business Objects Explorer, which provides most suitable functionality for the presentation of analytical data.
In order to compare the result of our work, we developed a benchmark that allows us to compare our prototypes with existing solutions, such as SAP’s Object Event Repository or Fosstrak EPCIS.
Parallel to the performance focus, the project implements and evaluates a Discovery Service aggregation algorithm. This new algorithm efficiently resolves containment relations between objects, reducing network traffic and processing time for Discovery Service queries.