Date: Wed Aug 3, 2016
Time: 8:00 AM - 10:00 AM
Moderator: George Vellidis
Most data sets for evaluating crop production practices have too few locations and years to create reliable probabilities from predictive analytical analyses for the success of the practices. Yield monitors on combines have the potential to enable networks of farmers in collaboration with scientists and farm advisors to collect sufficient data for calculation of more reliable guidelines for crop production showing the probabilities that new or existing practices will improve the efficiency of food production. The creation of sufficiently large data sets requires the pooling of data from numerous farmer networks, but such pooling of data is currently not possible because there are no standards for sharing of data across networks. The objectives of this paper are to: 1) provide a rationale for a community for on-farm data sharing; 2) describe the challenges of sharing data from on-farm networks and of sharing research data in general; and 3) identify the benefits of data sharing by reviewing what could be gained if data were shared across existing networks in the Corn Belt of the US. Writing and publishing standards for stewardship of data from farmer networks that would include standards for sharing and confidentiality of the data will encourage the creation of large data bases of results from replicated strip trials. The benefits from large data bases of such results are enormous. The greatest benefit is agronomists would be able to move away from the common practice of analysis that answers only the question whether there was a treatment effect to analyses that provide reliable probabilities of the chances a crop production practice will improve the efficiency of food production, and the magnitude of the treatment effect.
AgGateway is a nonprofit consortium of 240+ businesses working to promote, enable and expand eAgriculture. It provides a non-competitive collaborative environment, transparent funding and governance models, and anti-trust and intellectual property policies that guide and protect members’ contributions and implementations. AgGateway primarily focuses on implementing existing standards and collaborating with other organizations to extend them when necessary.
In 2010 AgGateway identified interoperability in field operations (e.g., planting, harvest) as a major challenge: the myriad proprietary, incompatible data formats among different machine and implement control systems (MICS) and farm management information systems (FMIS) leads to user frustration and operational inefficiencies.
The SPADE project was AgGateway’s first field operations effort. SPADE has generally followed an Agile Methodology for each field operation: collect expert-mediated user stories; formalize them into business process models and use cases; identify data requirements; perform gap analyses of existing standards, and propose extensions thereto. A key SPADE outcome was the identification of a set of flexibly-defined common "documents" that support farm business processes: Crop Plan, Observations and Measurements (O&M), Recommendation, Work Order, and Work Record.
SPADE identified Reference Data: datasets to distribute across the whole industry so different stakeholders can interpret shared documents the same way. This includes names and identifiers of seed varieties, crop protection products, active ingredients, etc. SPADE implemented proof-of-concept (PoC) application programming interfaces (APIs) for machinery and product Reference Data, and also implemented a PoC index API providing one-stop access to a distributed system of standardized Reference Data sources hosted by manufacturers and third-party data providers. Other SPADE contributions include defining concepts such as “OK to Spray”, contributing terms to AgGlossary.org, and working with the Agricultural Electronics Manufacturers’ Foundation (AEF) to expand the widely-respected ISO 11783-10 MICS-FMIS communication standard’s ability to associate globally-unique identifiers to its own locally-scoped identifiers.
CART is a SPADE sub-project. Its focus is to support grain movement and testing processes, aligning with (and expanding) the AgXML Consortium’s CommodityMovement and QualityCertificate messages.
WAVE is another SPADE sub-project. Its scope includes mobile and fixed asset management. It seeks alignment with the new ISO 15143-3 mobile asset telemetry standard. Its fixed asset scope (e.g., grain bin temperatures) also seeks existing-standards alignment, and dovetails with O&M work in PAIL and various SPADE teams.
PAIL is SPADE’s irrigation-specific sibling. Given the lack of existing irrigation data exchange standards, PAIL is proposing one (via ASABE) that documents irrigation operations and related O&M, including web service-mediated data-exchange schemas.
SPADE successfully explored the feasibility of an open-source format-conversion toolkit, leading to what is now ADAPT. ADAPT’s scope includes a common object model derived from SPADE / PAIL requirements including ISO 11783-10 compatibility, a format-conversion framework, and an architecture where manufacturer-specific “plugins” convert between proprietary formats and the common model. ADAPT supports FMIS-to-FMIS communication, including the nuanced business processes not covered in ISO 11783-10.
Internationalization is an important aspect of this work; several conflicting requirements (universality vs. geopolitical-context-specificity, and controlled vocabulary vs. extensibility) were reconciled creating a ContextItem class and drawing from AEF’s experience managing controlled vocabularies separately from the ISO 11783-10 data model.
Modus Standard is a system of defined terminology, agreed metadata and file transfer format that has grown from a need to exchange, merge and trend agricultural testing data. The three presenters will discuss steps taken to develop the system, benefits to data exchange, current user base and additions being made to the standard.
The agriculture sector is a unique sector due to its strategic importance for both citizens (consumers) and economy (regional and global), which ideally should make the whole sector a network of interacting organizations. The FOODIE project aims at building an open and interoperable agricultural specialized platform hub on the cloud for the management of spatial and non-spatial data relevant for farming production. The FOODIE service platform deals with including their thematic, spatial, and temporal characteristics as well as their meta-information. Most of current information systems deal with data in the proprietary structure that fits the most to the purposes of each organization or individual. The structure of data itself is commonly proprietary even if there is a standard exchangeable format used by an information system. As a result, many organizations as well as individuals produce massive amounts of information for internal and external purposes. If we would like to integrate data from several sources, we need to establish a unified data model that is capable to include all the nuances of the underlying data models. In this regard, the conceptual model of FOODIE relies upon the following basic pillars: Data and service modelling in the geospatial domain relies upon the series of ISO/OGC geospatial standards and best-practices, Data and service modelling in the agriculture domain relies upon specific agriculture standards and best-practices such as the INSPIRE data model for Agricultural and Aquaculture Facilities, Transport and Monitoring Facilities. The paper described data models implemented as part of FOODIE infrastructure. In addition, FOODIE data model specification ensures: unified understanding and concept of agricultural data; interoperability with other similar (sub)systems when using open data in a standardized way; compliance with existing legislative requirements in Europe (especially INSPIRE and LPIS).