agriopenlink @Precision Dairy Farming 2015 (Rochester, MN)

SEMANTIC TECHNOLOGIES IN THE INFORMATION MANAGEMENT

FOR PRECISION DAIRY FARMING: A SHOWCASE OF AGRIOPENLINK

Dana Tomic, Dr. (FTW), Wolfgang Auer (MKWE), Sandra Hoermann(JR)

Dana Tomic, PhD (FTW)

PRECISION DAIRY FARMING, Mayo Center, ROCHESTER, MIN, 24-24 June 2015.

agriOpenLink Key Facts

Interdisciplinary Research Project

- FTW - Research Center for ICT

- JR - Josephinum Research

- BOKU - University of Natural Sciences BOKU

- MKWE - Animal Monitoring System (SMARTBOW)

The main theme: Information Management for Processes Optimization in Precision Agriculture

Funded by: Austrian Research Funding Agency (FFG)

Use Cases: 1) Dairy Farming, 2) Irrigation

Duration: 3 Years ( 06’2013- 05’2016)

Budget: ~900 k Euro (80% funding)

2


Precision Dairy Farming

A vibrant field of innovation!

The real-world data and information

processing enables new management

practices!

Machines are substituting & supplementing

people not only in in milking and feeding

tasks, animal and herd monitoring but also in

making decisions !

To make better decisions we need better

- Data and knowledge integration

- Interoperability

- Knowledge actuation

3


agriOpenLink Vision


SMARTBOW - Based on ear tags.

• Activity / resting

• Fertility

• Rumination

• etc.

- Improved nutrition and

fertility management,

animal wellbeing and

health.

80 000 per cow per day

Color codes for activities (resting, feeding, moving)


Insights: Farmers

..

How to quantify add-value for a PDF system?

- A single solution is not sufficient for effective advice!

An integration concept is needed !! 6

Bewley, 2013


Cond.

Protein

Resting

Yield

Milking

Insights: Researchers

Researchers are early adopters of the PDF

- Testing new models

Challenges

- How to obtain data needed to verify models for different farm configurations and constraints

- How to translate verified models / knowledge

into operational systems for farmers

Rutten et al, 2013, Sensors to support health management … 126 pubs. /139 systems

- 4 Level Model of functionality

- mastitis (25%), fertility (33%), locomotion (30%), metabolism (16%)

- No Level 3&4 systems, for level 1&2 the detection performance, the selection of indicators, sensor techniques, and gold standards require further work

7

Level 1:

Measurement

Level 2:

Interpretation

Level 3: Integration

Level 4:

Automation/DS


Insights: Vendors

Systems tightly integrated using mostly proprietary interfaces and data models

- high entrance barrier for Level 1&2 systems / high costs of updates

Interface standards

- ISOagriNET ADIS/ADED (www.isoagrinet.org)

- ISO17532: Data comm. Network, ISO 11787, Data interchange syntax (ADIS), ISO 11788 Agricultural data element dictionary (ADED)

- ASCI files - Optimized for transmission

- Do not allow for fast changes

- Isolated information island

- You can create a readable report but Is not meant to encode knowledge or integrate data

- More and more also use Web Service standards (internally)

Providing consulting to farmers is very important! - Knowledge management approach may be beneficial

- Eastwood et al., 2009, Farmers as co-developers …

8


Knowledge Management Systems

KMS integrate:

- Domain knowledge

- Search

- Sharing of knowledge

- Turning knowledge into actions

Embrace the Semantic Web Technology

The Semantic Web is an extension of the current web in which information is given well-defined meaning (formally described knowledge), better enabling computers and people to work in co-operation (Berners-Lee et al., 2001)

- Considerable maturity of tools and standards (Auer, 2014).

Domains of adoption: open government solutions, corporate information systems in eHealth, tourism, media publishing, …

9


Formal Description of Knowledge

Make it possible for machines to process the data and create new knowledge!

Description languages (W3C Consortium)- RDF and RDF Schemas

- Simple Knowledge Organization System (SKOS)

- Web Ontology Language (OWL)

Ontology = explicit formal specifications of the terms in the domain and relations among them (Gruber, 1995)

- Classes, relationships, data properties, instances

- Primitive classes for basic hierarchy: Animal, Organization, Location, and properties (configured or observed parameters from different sources)

- Defined classes for classification

Ontologies are linked with other ontologies and are published on the Internet!


Interoperability via Web Services

Web Services

- Aiming to provide interoperability among distributed

loosely coupled components

- Invoked by using standard Web protocols, SOAP, HTTP,

REST

- Use machine-interpretable descriptions of services to

automate: discovery, invocation, composition and

monitoring of

- Web agents can compose simple web services into

complex web services

Semantic Web Services

- Enriched with the semantic metadata (based on ontology)

- Use and generate structured (semantic data)

- Dynamically chained in flexible workflows based on

reasoning on data

11


Knowledge Actuation via Semantic Web Services

12

Everything

arranged


agriOpenLink Platform

Semantic Web approach is used for its unique flexibility in knowledge description, knowledge sharing, system design integration and configuration

Technology for

- Incremental inclusion of systems

- Data and knowledge sharing

- Knowledge-based collaborationamong the dairy producers, experts and system vendors.

- Decision support & process automation

Lowering barrier for all data publishers

- Multi-vendor systems

- Emerging health monitoring systems

13

Plugin

Semantic

Web Services

Ontology

Queries,

Level 1:

Measurement

Level 2:

Interpretation

Level 3: Integration

Level 4:

Automation/DS


agriOpenLink Technical Realization

14

Plugin Component

- Plugin Gateway

- Plugin Server (HTTP REST+RDF)

- Plugins (loadable components)

- Plugin Services (RDF)

Query Component

- Query Editor and EngineConsultant

Farmer

Plugin

Developer

Ontology Component

- Dairy Farming Ontology

- Repository with Ontology & Data

- DFO Editor for collaborative editing


Knowledge Actuation via Semantic Web Services

15

Which cows had a milk yield

above average within the last

month, and had alarm

condition that requires

medical attention

To Do:- Check the definition of alarm

condition, medical attention

- Check the milking system

- Check monitoring system

- …

Bob informed !Lucy & Ilse will be checked

Activity data show anomaly …


agriOpenLink @EuroTier


Current Work

17

Test Installation:

- Farm Edermaier, 80 ~Cows

- St. Peter in der Au (NÖ)


Conclusions and Outlook

Our vision is possible to reach!

Plugin Environment reduces the cost of

integrating new systems!

DFO is a basis for establishing a shared

integration context for data and knowledge

integration, for recommendations and

encoding and sharing knowledge.

Next Steps:

- Support for interested system vendors in the

design of their own plugins

- Experimenting with the DFO extension

- Experimenting with on-the-top applications

18

Cost of the integration

traditional

Cost of the integration

agriOpenLink

$

number of systems


Contact

Dr. Slobodanka Dana Kathrin TomicSenior Researcher | FTW | www.ftw.atForschungszentrum Telekommunikation Wien GmbHDonau-City-Straße 1/3 | A-1220 Vienna | Austria+43/1/5052830 -54 | fax -99 | +43/6769129023

agriopenlink @Precision Dairy Farming 2015 (Rochester, MN)

Food