A Taxonomy of the Data Resource in the Networked Industry

© Fraunhofer
TOWARD A TAXONOMY OF THE DATA
RESOURCE IN THE NETWORKED INDUSTRY
Boris Otto, Rene Abraham, Simon Schlosser
Cologne, June 5, 2014

© Fraunhofer
AGENDA
 Data in the Networked Industry
 Research Approach
 Case Studies on Data in the Networked Industry
 Data Morphology Design
 Method Support
 Outlook

© Fraunhofer
A set of current developments foster the adoption of
networked forms of organization in many industries
Globalization
Internet of Things
Consumer-Centricity
Product Complexity
Networked
Forms of
Organization

© Fraunhofer
The role of data has evolved from a by-product to a
product in its own right traded on data markets
Factual InfoChimps Windows Azure
Data Market
Data.com
Year of
Foundation
2007 2009 2010 2010 (formerly
Jigsaw, 2004)
Owner Venture capital
firms
CSC Microsoft Salesforce.com
Offering Open data
platform, API use
for free or at a
charge.
15,000 data sets,
open data platform,
four
different pricing
models, web service.
Wide range of data,
including open data
platform. Buying and
selling data via
Azure marketplace.
Data sets for
increasing master
data quality,
maintained by
community of
2.000.000 users.
Services Data mining, data
retrieval, data
acquisition from
external parties.
Data collection,
infrastructure
development,
hosting and
distribution.
Software as a Service
(SaaS) applications
and data sets,
partially real-time
access.
Different service and
pricing models.
Access to contact
information, real-
time updated data
sets.

© Fraunhofer
Companies in the networked industry struggle with
finding an appropriate data architecture
Data in the outer circles is of higher
“fuzziness”, volume, change frequency…
Data in the outer circles is of less
control, criticality, unambiguity…
“Nucleus Data”
(Customer master
data, product
master data etc.)
“Community
Data”
(Geo-information,
GTIN, addresses,
ISO codes, GS1
data etc.)
“Open Big Data”
(Tweets, social media
streams, sensor data
etc.)
Megabytes
Gigabytes
Terabytes
Petabytes

© Fraunhofer
The scientific knowledge base falls short in explaining
the role of data in the networked industry
Networked Industry
Perspective
Selected Contributions with Data Focus Summary of Knowledge Base
Enterprise (Addo-Tenkorang, Helo, Shamsuzzoha, Ehrs &
Phuong, 2012), (Bettoni, Alge, Rovere,
Pedrazzoli & Canetta, 2012), (Legner &
Schemm, 2008)
Data modeling in supply chains
Supply chain data management
Network (Howard, Vidgen, Powell & Graves, 2001),
(Lampathaki, Mouzakitis, Gionis, Charalabidis &
Askounis, 2009), (Legner & Schemm, 2008),
(Nelson, Shaw & Qualls, 2005)
Data and information sharing
Data standards
Interoperability
Technology (Chalasani & Boppana, 2007), (D'Amours,
Lefrançois & Montreuil, 1996), (Derakshan et
al., 2007), (Dreibelbis et al., 2008), (Parlanti,
Paganelli & Giuli, 2011) (Wang & Jin, 2008)
Data as a service (SOA)
Information systems design
RFID data architecture design

© Fraunhofer
The goal to increase understanding of data in the
networked industry translates in two research questions
Research Question 1
 How does a morphology of the data resource in the networked industry
look like?
Research Question 2
 How should a methodology be designed that helps companies in the
networked industry to apply the morphology for data architecture
design?

© Fraunhofer
The explorative and design-oriented approach follows a
two-phased research process
Phase IIPhase I
Literature
Review:
DRM/DAM
Case Analysis
Morphology
Analysis and
Design
Literature
Review: DRM
Method
Engineering
Method for
Morphology
Application
Legend: DRM - Data Resource Management; DAM - Data Architecture Management.

© Fraunhofer
Four cases were analyzed for morphology analysis and
design
Case A B C D
Perspective Consumer-Centricity Supply Chain
Excellence, IoT
Purchasing Electronic
commerce
Industry Consumer goods
and retail
Consumer goods
and retail
Pharmaceutical,
chemical, food
Online retailing
Data objects in
focus
Suppliers, retailers,
products, consumers
Suppliers, retailers,
load carrier
Suppliers Customers,
products
Case study partners Beiersdorf, Migros Mars, Rewe, Chep Bayer, Nestlé,
Novartis, Syngenta
Amazon
Data collection and
analysis
Interviews
Participatory case
study
Expert interviews
Case study
Interviews, focus
groups, data
overlap analysis
Participatory case
study
Archival records,
public
documentation
Case Study
Project context Competence Center
Corporate Data
Quality
SmaRTI Corporate Data
League
-

© Fraunhofer
In Case A, Beiersdorf analyzed the betweenness of
product data flows in its network
Agency
Consumer information provider
Brand
owner
Consumer
Retailer
Consumer
Agency
Consumer
information provider
Consumer technology
provider
GDSN
Social
network
Online
retailer
Brand
owner
Retailer
web shop
Forum &
Blogs
2007 2012
Legend: GDSN - Global Data Synchronisation Network.
Media

© Fraunhofer
Analysis of Case A revealed shortcomings when it comes
to managing data in a networked industry
 Today, the label drives product data management
 Carbon foot print information or allergen implications not considered
 Product data quality differs
 High quality in supply chain data, low quality with regard to product
information
 Data sources are not transparent when controlled by the consumer
(ratings, blogs, posts about products etc.).
 Variety of data formats increases (videos, streams, images etc.)

© Fraunhofer
Case B analyzes the consumer goods supply chain in the
context of the SmaRTI project
Cloud-based data service for data aggregation
and provisioning etc.
 Cloud-based
 Service-oriented
 Standardized
Intelligent load carriers such as
 Retail pallets
 Air cargo pallets
Process modeling following Internet
of Things design principles
 Self-controlled
 Decentralized
Internet of Service
 Data marketplace
 Business intelligence
 Apps

© Fraunhofer
Analysis of Case B revealed shortcomings when it comes
to managing data in a networked industry
 Collaborative environment needed to collect, aggregate, analyze data
from EPCIS events
 Value network-wide standardization of data formats and semantics
needed
 Traditional design principles for application systems becoming obsolete
 Maintaining pallets as stock items
 Real-time data availability on item level conflicts with standard
document flow
 Ownership of collaborative data unclear
 Integration of structured ECPIS data and value-added PoS and
multimedia data not clear
Legend: EPCIS - Electronic Product Code Information Services; PoS - Point-of-Sale.

© Fraunhofer
The data morphology for the networked industry covers
various dimensions
Dimension Characteristics
Business criticality Competitive advantage Compliance relevant Operations relevant
Data classification Private Public Purpose-related
Data domain type Account Party Thing Other
Data format ASCII Audio JPEG Video Numeric XML
Data management level Class Instantiation
Data occurrence Batch Stream
Data ownership Owned by one legal entity “Club” good Public good
Data quality Authoritative Within tolerance, fuzzy Below thresholds
Data source Internal External
Data standardization Semantics Syntax Values
Data trustworthiness Not trusted Trusted
Data sharing Open Free Proprietary
Data maintenance costs Low Medium High

© Fraunhofer
Phase I: Identify domain and scope
A method provides methodological support for applying
the morphology in practice
 Design data architecture
 Create transparency
 Managing risks
 Find data management
patterns
Activities Results Roles
I.1 Define scope
I.2 Identify data
objects and items
Phase III: Design
Phase II: Analyze
II.1 Create
transparency
II.2 Analyze
and assess
III.1 Derive design
requirements
III.1 Design data
architecture
Identified data domain
and analysis objective
List of data objects and
items to be analyzed
Data steward
Data steward, data
architect, data
owner
Data steward, data
owner, data
scientist, (business
partners)
Data scientist, data
architect
Data (heat) map
Risks and opportunities
Requirements list
Data architecture
Data architect, data
steward
Data architect

© Fraunhofer
The morphology identifies data resource patterns as the
example of business partner data from Case C shows
Dimension Characteristics
Business criticality Competitive advantage Compliance relevant Operations relevant
Data classification Private Public Purpose-related
Data domain type Account Party Thing Other
Data format ASCII Audio JPEG Video Numeric XML
Data management level Class Instantiation
Data occurrence Batch Stream
Data ownership Owned by one legal entity “Club” good Public good
Data quality Authoritative Within tolerance, fuzzy Below thresholds
Data source Internal External
Data standardization Semantics Syntax Values
Data trustworthiness Not trusted Trusted
Data sharing Open Free Proprietary
Data maintenance costs Low Medium High
Legend: The darker the more apprproiate.

© Fraunhofer
The research has limitations and points the ways to
some further research opportunities
 Limitations
 Qualitative data
 First design cycle only
 Morphology needs refinement
 No large scale evaluation
 For pattern detection
 Outlook
 Data architecture patterns for verticals
 Elaboration of methodological support
 Networked data management systems

© Fraunhofer
Please get in touch for further information
Univ.-Prof. Dr. Ing. habil. Boris Otto
TU Dortmund University
Audi-Endowed Chair of
Supply Net Order Management
LogistikCampus
Joseph-v.-Fraunhofer-Straße 2-4
D-44227 Dortmund
Tel.: +49-231-755-5959
Boris.Otto@tu-dortmund.de
Fraunhofer Institute for
Material Flow and Logistics
Director Information Management
& Engineering
Joseph-v.-Fraunhofer-Straße 2-4
D-44227 Dortmund
Tel.: +49-231-9743-655
Boris.Otto@iml.fraunhofer.de

A Taxonomy of the Data Resource in the Networked Industry

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Viewers also liked

Viewers also liked (11)

Similar to A Taxonomy of the Data Resource in the Networked Industry

Similar to A Taxonomy of the Data Resource in the Networked Industry (20)

More from Boris Otto

More from Boris Otto (6)

Recently uploaded

Recently uploaded (20)

A Taxonomy of the Data Resource in the Networked Industry