Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Fairtrace - Tracing the textile industry


Published on

A l’occasion de la 11ème conférence internationale sur le Web sémantique (ISWC 2012) qui a eu lieu du 11 au 15 novembre 2012 à Boston (USA), l’Institut informatique de gestion de la HES-SO Valais a présenté FairTrace, une solution permettant la traçabilité de la production de vêtements au moyen d’applications mobiles. Ce projet représente une véritable avancée technologique reconnue au niveau mondial mais également une réussite économique. En effet, en octobre dernier FairTrace est devenue une start-up de l’incubateur The Ark à Sierre!

Published in: Technology
  • Be the first to comment

  • Be the first to like this

Fairtrace - Tracing the textile industry

  1. 1. FairtraceTracing the textile industry Bruno Alves University of Applied Sciences Western Switzerland ISCW2012, Boston 14th Nov. 2012
  2. 2. 2Traceability concerns • Euro2012 Fan’s Shirts • Greenpeace Germany commissioned report • REACH regulation No or partial knowledge of their supply chain
  3. 3. 3Fairtrace
  4. 4. 4 Technology Operative BackendB2B Frontend B2C Frontend Operative Frontend
  5. 5. 5Little demo DEMO /ˈdɛm.əʊ/
  6. 6. 6Little demo − Formulars
  7. 7. 7Little demo − Datasets DEMO /ˈdɛm.əʊ/
  8. 8. Understanding the textile 8process Question: How do we define our models? • Document search • Link chaining • Minimal data set
  9. 9. 9Defining the data model(s) RDF Schema vs OWL*
  10. 10. Implementation − problems & 10 techniques• Data depth & query complexity – Gather data at single points• Link formulars to data model ; ?dataset fpos:hasDatasetUser ?user Id: fto5 – fpos:hasDatasetForm ?form ; Data binding and instantiation y ; x <ftos:hasProcessedFabricOrder> fpos:hasDatasetScope <.../order001> x <ftos:hasProcessedFabricColor> z mechanisms fpos:hasDatasetTimestamp ?timestamp . ------------------------------- y <ftos:hasOrderColor> z• Data scopeOrder#buyer_code → fpos:hasOrderBuyerCode – Order subclassOf Scope, Dataset
  11. 11. 11Testing • Tee-shirt • Consultant in India • No metrics • Issues were solved Positive feedback
  12. 12. 12 Competitive advantage• No dependency on the schema – Greater adaptability• Different product domains – Change product ontology• Easy data binding – Different process / products
  13. 13. 13 Lessons learnt• Inferrences can be confusing – subClassOf• Consistent naming conventions – hasOrderId vs hasId• Data update and deletion is scary – is data still there ?
  14. 14. 14 Conclusions• Traceability becomes a trend• Start-up (CTI-Startup) with a commercializable product – Patent pending – Already one customer• Linked Data / SPARQL 1.1• Research
  15. 15. 15Thank you ! Bruno Alves Senior Scientific Collaborator Email: Phone: +41276069034
  16. 16. EXTRA: Choosing the 16 storage technologyChoice: Relational databaseor semantic store ?• Query• Depth• Missing information