big data and semantic web in manufacturing
DESCRIPTION
Big Data and Semantic Web in ManufacturingTRANSCRIPT
Big Data and Semantic Web in Manufacturing
Nitesh Khilwani, PhD
2
Outline
• Big data in Manufacturing• Big data Analytics• Semantic web technologies• Case study:
– 1. ECOS Ontology– 2. Text2Rdf– 3. Semantic Layer for Information management.
3
Manufacturing
• Gone are the days when a manufacturing companies operated autonomously
• Using advanced IT systems companies are linking with consumer, manufacturing operations and suppliers to work as single entity.
MINING/FARMING SUPPLY CHAIN MANUFACTURING DISTRIBUTION CUSTOMER
4
Manufacturing
• Lots of data is being collected and warehoused – purchases at department/
grocery stores– Bank/Credit Card
transactions– Social Network– Web data, e-commerce
• Big data: A collection of data sets so large and complex that it becomes difficult to process using on-hand database management tools.
MANUFACTURING
Business
Technology Workforce
5
Big Data in Manufacturing
• Big Data in manufacturing has the potential to revolutionize how we build, produce and live.
• For decades, manufacturers have invested in and developed technologies that leverage “just in time” and “Six Sigma” methodologies.
• Next step is data analytics...
Big Data:Decisions based on
all your data
Video and Images
Machine-Generated DataSocial Data
Documents
6
Big data Use cases
Today’s Challenge New Data What’s Possible
ManufacturingInter department Support Product sensors Automated diagnosis, support
HealthcareExpensive office visits Remote patient monitoring Preventive care, reduced
hospitalization
Location-Based ServicesBased on home zip code Real time location data Geo-advertising, traffic, local
search
Public SectorStandardized services Citizen surveys Tailored services,
cost reductions
RetailOne size fits all marketing Social media Sentiment analysis
segmentation
Big Data Is About…
Tapping into diverse data sets
Finding and monetizingunknown relationships
Data driven business decisions
8
Big Data Landscape
9
Tools for Big data Analytics
• Visualization: baseline tools for both experienced data scientists and more novice analysts to make sense of data.
• Data mining: To make machines more capable to automate the analysis.
• Predictive Analysis: Tools to look forward and analyse the problems
• Semantic engine: Tools to parse, extract, and analyze data
• Data quality and Master Management: To insure quality and management process around data.
10
Gartner Life Cycle of Emerging Technologies -2013
• Big data and supporting technologies are part of Hype Cycle
11
Big data in Manufacturing: Benefits and Challenges
Benefits
Product quality and defect trackingSupply chain management
Forecasting of manufacturing output
Increase efficiency
Simulation and testing of new processes
Enabling customization in manufacturing
Challenges
Building trust between data scientist and managers.Confidence to handle large volume, velocity and variety of dataFinding the door for Big data to enter our current system.Determining which technology to useDetermining what to do with the analysis from Big dataMaintaining the consistency for using big data.
12
Information floating in Manufacturing
Bug Tracking and Management
(Bugzilla)
Version Control (SVN, Perforce)
Knowledge Management System (Wiki)
• Issue• Project• Component• Release• Status• Tester• Developer
• Component• Code• Version• Issue• Developer• Activity
• Project• Requirement• Milestones• Test plan• Process• Author
Competency Management
(Website)
• Hierarchy• Organizational • Structure• Roles• Responsibility• Developer• Tester
13
Information floating in Manufacturing
Bug Tracking and Management
(Bugzilla)
Version Control (SVN, Perforce)
Knowledge Management System (Wiki)
• Issue• Project• Component• Release• Status• Tester• Developer
• Component• Code• Version• Issue• Developer• Activity
• Project• Requirement• Milestones• Test plan• Process• Author
Competency Management
(Website)
• Hierarchy• Organizational • Structure• Roles• Responsibility• Developer• Tester
14
Information floating in Manufacturing
Bug Tracking and Management
(Bugzilla)
Version Control (SVN, Perforce)
Knowledge Management System (Wiki)
• Issue• Project• Component• Release• Status• Tester• Developer
• Component• Code• Version• Issue• Developer• Activity
• Project• Requirement• Milestones• Test plan• Process• Author
Competency Management
(Website)
• Hierarchy• Organizational • Structure• Roles• Responsibility• Developer• Tester
• Interval between bug fixing time and change commit time
• Mapping between bug owner and change committer
• Similarity between bug report and change logs
• Alternate contact points for emergency situations.
15
Architecture for Data Analysis
CRM
Applications
APPLICATION FRAMEWORK
Authentication, Authorization, Query transformation, Personalization,
display
Applications Applications
DATA CONNECTION
LAYER
APPLICATION LAYER
USER MANAGEMENT
LAYER
PROCESSING LAYER
SEARCH ENGINE
IndexingCrawling
Converting
TEXT ANALYTICS
ThesaurusClustering
Semantic ProcessingEntity Extraction
SEMANTICS
MetadataOntologyTagging
Semantic Web
17
Semantic Web
• Semantic web is an extension of current web technology in which information is annexed with well defined meaning– Provides machine processable
semantics of data
• Ontology: Basic building block for Semantic Web– Captures semantics and relations of
a domain
18
Semantic Web Technologies
• RDF: Language for representing knowledge
• RDFS: Vocabulary for defining classes and properties
• OWL: Adding relations in RDFS• SPARQL: Query language based
on graphs.• Rules: Language to combine
different ontology.
19
Ontology: ECOS
• ECOS Ontology: Enterprise Competence Organization Schema
– Publishing enterprise competence in an explicit and structured manner
20
ECOS
• Use other ontologies to extend the ECOS.• Standards defined by UN and EU are used in ECOS for representing
competences
General
Specific
Enterprise RecordBusiness
Company Name Contact PersonAddress Key Person
Past Projects
Relation
Resource Process Unit Skill
Product/Service
Customer
Sector
Preference
Financial
Summary
Plan
Achievement
21
TEXT2RDF
TEXT2RDF
Text mining approach for providing explicit semantic description to the
documents.
22
Semantic Information in Manufacturing
Bug Tracking and Management
(Bugzilla)
Version Control (SVN, Perforce)
Knowledge Management System (Wiki)
• Issue• Project• Component• Release• Status• Tester• Developer
• Component• Code• Version• Issue• Developer• Activity
• Project• Requirement• Milestones• Test plan• Process• Author
Competency Management
(Website)
• Hierarchy• Organizational • Structure• Roles• Responsibility• Developer• Tester
METADATA METADATA METADATA METADATA
SEMANTIC LAYER
23
Semantics in Enterprise
• Supply Chain Management—Biogen Idec• Media Management—BBC• Data Integration in Oil & Gas—Chevron• Web Search and Ecommerce
24