Ontology Summit 2013: Ontology Evaluation Across the Ontology Lifecyle
Using Ontology to Capture Supply Chain Code Halos
18
Using Ontology to Capture Supply Chain Code Halos Manufacturers need to create a lingua franca that extends throughout the supply chain ecosystem, in order to generate insights from the digital data encircling their employees, partners, processes and customers.
Transcript of Using Ontology to Capture Supply Chain Code Halos
Using Ontology to Capture Supply Chain Code HalosManufacturers need to create a lingua franca that extends throughout the supply chain ecosystem, in order to generate insights from the digital data encircling their employees, partners, processes and customers.
2 KEEP CHALLENGING November 20142 KEEP CHALLENGING November 2014
Executive SummaryThe world of high-end digital devices, with its rich hyper-connectivity,
portability and virtual intelligence, has become deeply engrained in
our professional and personal lives in ways heretofore unfathomable.
Furthermore, with the mainstreaming of technologies such as social
media and cloud, our worlds have grown both more complex and
interconnected.
While organizations are tapping into these technologies to generate
business value that extends beyond mere transactions, they are not
finding it easy to do so. Leveraging the metadata contained within these
digits offers unprecedented insight into the spirit and intent of these
transactions, beyond what is superficially apparent. However, the ability
to collect and distill insight from this data and generate exponential
value requires the ability to read what is hidden between the lines and
then structure it in ways that are understandable and useful to the
business.
Since the dawn of modern science, humans have aspired to understand
how the brain works and to replicate it through technology. This has
given rise to the discipline of artificial intelligence (AI), which today
holds the answer to how society can best capture intangible information
and structure it in a way that we can make sense of it.
This white paper explores ways to connect the dots presented by these
concepts. We provide a foundation to build capabilities specific to the
supply chain management world to produce groundbreaking insights
from the data that is generated every day.
USING ONTOLOGY TO CAPTURE SUPPLY CHAIN CODE HALOS 3 USING ONTOLOGY TO CAPTURE SUPPLY CHAIN CODE HALOS 3
Data Rich, Knowledge PoorExabytes of digital data (1,000 petabytes)1 are generated around the world every day. Within a supply chain, data is generated by the activities of the supply chain players, as well as at the points of interaction among these players. This phenomenal amount of digital data encircling individuals, companies, processes and devices is what we call a Code HaloTM. By making meaning from this data, supply chain constituents can generate immense value for themselves and their partners.2
To do this, organizations must make meaning from unstructured or partly structured data, using a comprehensive representation of what we know about the data and its interrelation-ships. Existing models such as SCOR (Supply Chain Operations Reference) from the Supply Chain Council have achieved a consensus view of supply chain management from a business process perspective. These models have helped supply chain managers establish metrics, standardize language and create common business practices. This activity has also led to academic research to extend the SCOR model and combine it with knowledge representation through an ontology known as SCONTO.3
Ontology is “a formal, explicit specification of a shared conceptualization,”4 or in simpler terms, a hierarchical description of concepts in a certain domain, combined with a description of each of these concepts.5 By applying ontology, Code Halos associated with the supply chain ecosystem can be codified and represented by providing a formal structure to the data on the basis of our knowledge about it.
This information can then be extended to specialized application areas within the supply chain and captured across the various information management systems used by players in the supply chain. This will result in an information base that is easy to integrate across supply chain partners and is reusable by all participants.
For example, ontologies are widely used in e-commerce to represent accurate product informa-tion, specifications and hierarchy across a wide range of products, categories, partners and Web sites.
However, it is not always necessary to fit businesses to standard models when using ontology to make meaning of unstructured or partly structured data. Also, businesses that have adopted standard models like SCOR have often needed to customize the model to different degrees to make it less ambiguous, more robust, and a better fit for their needs. Businesses that do not follow standard academic models (which are the majority of cases) can leverage the power of Code Halos through ontology, using the approach suggested in this paper.
Supply Chain Code Halos The digital data surrounding entities in a supply chain can be classified in a number of categories, largely driven by the context of the data within the broader business ecosystem. The value generated by this data depends on the ability to identify the most effective intent of using it rather than the pure transactional nature of the information it communicates (see Figure 1, next page).
4 KEEP CHALLENGING November 2014
Businesses that do not follow standard academic models can leverage the power of Code Halos through ontology, using the approach suggested in this paper.
Manufacturers and PartnersManufacturers generate massive amounts of data, both from transactional sources and through interactions among multiple systems and individuals. Chief among this data are the metrics and statistical data generated by manufacturing processes. This data originates from multiple sources within the core manufacturing environment, for example, the control system components of a process-based manufacturing execution system. With the adoption of mobility and close integra-tion with enterprise resource planning (ERP) systems, the volume, velocity and variety of data has also increased substantially.
Moreover, data on best practices and standards is spawned from collaboration platforms used to share expertise, experiences and perspectives (e.g., Yammer, Chatter, etc.). This can be further enhanced by sourcing data from external forums to obtain a more holistic view of best practices and standards of manufacturing processes.
USING ONTOLOGY TO CAPTURE SUPPLY CHAIN CODE HALOS 5
Figure 1
Code Halos in the Supply Chain
Products and Services Data generated from PLM systems, product reviews, customer feedback, collaborative platforms, network-aware devices, field services, bill of materials, etc.
End Customer Data generated by social networks, search engines, e-commerce systems, online forums, transaction history, surveys, customer service, payment systems, etc.
Transportation and Logistics Data generated from live traffic and weather updates; GPS and satellite-based systems; government entities; order fulfillment, real-time inventory and returns authorization systems, etc.
Manufacturers and Partners Data generated from manufacturing processes; best practices information from collaboration platforms and online forums; competitive information from the Web and paid sources, etc.
6 KEEP CHALLENGING November 2014
Collaboration platforms with suppliers, distributors, vendors and other third parties can prove to be another source of useful business data. For example, supplier portals to interchange ASNs, EDI channels, etc. can provide useful insights on the efficiency of the partner operations and other inputs beyond the purely transactional.
Moreover, another source is derived from access to competitor information and intelligence from open channels on the Web. Subscriptions to various paid channels of competitor informa-tion (e.g., Hoovers) can potentially be added to this.
Products and ServicesProducts and services — which are supply chain outputs — are a significant source of useful business data. This covers both data generated by the products themselves through the product lifecycle management systems, as well as the data about products generated from external sources, such as product reviews and feedback from e-commerce Web sites or feedback forms.
Collaborative product design and development provides a host of useful business data from various partners that can offer fresh perspectives. Hence, collaborative product development platforms — part of modern-day product lifecycle management systems — are also a rich source of this kind of product data.
Network-aware devices and products that make up the Internet of Things (IOT) are generating exponential amounts of data that can provide useful insights about both usage and performance. For example, smart meters attached to electric grids yield insights on a customer’s power usage patterns and preferences beyond the basic information of the grid’s operational state.
The above can be combined with more traditional product data, such as bill of materials and component sourcing information, to provide a holistic view of products. The downstream supply chain data from field services, such as servicing requirements, servicing frequency and usage patterns, can also add more traditional but very useful sources of product data.
Transportation and LogisticsData sources for logistics operators have vastly increased. Whether it’s live traffic updates and weather information, or vehicle positioning and tracking using GPS and other satellite-based systems, a huge amount of data is generated by various commercial and public systems. Businesses that can tap into these sources using telematics-based technologies can make effective logistics decisions in real-time involving route planning, fleet management, safety, accident prevention, etc. Moreover, government data sources are in abundance, such as facilities to verify drivers’ licenses by departments of motor vehicles throughout the U.S. and UK.
Collaborative product design and development provides a host of useful business data from various
partners that can offer fresh perspectives. Hence, collaborative product development platforms — part of modern-day product lifecycle management systems —
are also a rich source of this kind of product data.
USING ONTOLOGY TO CAPTURE SUPPLY CHAIN CODE HALOS 7
From an execution point of view, data pertaining to real-time order fulfillment statistics and proof of delivery tracking, combined with real-time inventory information from back-end systems, can provide a live view of fulfillment and useful insights on performance and potential problems. Fur-thermore, businesses can track returns data through return material authorizations, combined with order and fulfilment data to potentially perform root-cause analysis of the causes for returns, thus reducing returns frequency.
End CustomerNew sources of customer data are providing groundbreaking insights into customer behavior, explicit and implicit expectations, usage patterns, etc. The vast amount of data generated in social networks, search engines, e-commerce systems, online forums, etc. through devices of different form factors — from computers to phones, watches and fitness gear — has immense potential to provide insights into the needs, wants and desires of existing and potential customers.
Insights on potential customers can be generated by sources such as search patterns, buying histories, social network updates (including “likes”), page visits and even buying patterns of related products or services. Insights into existing customers can come from satisfaction surveys, service history, reviews, feedback, discussion forums, payment systems, etc. For example, a global vehicle manufacturer created an early warning system by tapping into automotive Web sites and forums using Web “crawlers” to gather insights on what customers were talking about regarding newly launched product lines. Based on these insights, this company was able to more quickly address and manage possible issues before they became widespread.
Ontology to Make Meaning from Code HalosOnce businesses understand how many forms and sources of data are available to continuously enrich supply chain Code Halos, they realize that they need a way to structure and represent these massive amounts of data to derive meaningful insights. Ontology can be an excellent way to provide a method to this madness and meaningfully represent what would otherwise look like chaotic data.
Ontology involves various concepts of classes, subclasses, inheritance, relations, properties, instances, etc. Practitioners who are familiar with object-oriented design concepts will find many of these concepts familiar. The most significant concept, however, is that of classes, their hierarchy and how they can be interwoven to represent related real-world entities. Once these concepts are clear, decision-makers use ontology to represent businesses, organizations and entities in a structured and streamlined manner.
Once businesses understand how many forms and sources of data are available to continuously enrich supply chain Code Halos, they realize that they need a way to structure and represent these massive amounts of data to derive meaningful insights.
8 KEEP CHALLENGING November 2014
Process RFI/RFQ*
Plan*
Source*
Make*
Deliver
Return*
Invoice*
Metric
Measures Supports
Subclass of
Subclass of
Subc
lass
of
Subclass
of
Subclass of
Subclass of
Definition
Make Cycle Time
Instance of
Der
ives
Instan
ce of
Instance
of
Instance of Derives
Derives
Best Practice*
Process Payment*
Deliver Stocked
Receive Order*
Deliver Made-to-Order*
Calculation
Deliver Engineered-to-Order*
Return on Working Capital*
Deliver Cycle Time
Source Cycle Time
Fulfillment Cycle Time
Upper Ontology
Computer
Desktop* Laptop
Handheld*
Upper Ontology
Phone*
Model XXX Model YYY
Specifications*
Subclass of
Subclass of
Subclass of
Instance of
Use
s
Contains
Bill of Materials*
Ultraportable Laptops*
Desktop Replacements*
Gaming Laptops*
Multimedia laptops
Two main types of ontology are useful in the context of supply chain information:
• Domain ontology represents knowledge in a particular domain.
• Process/task ontology represents knowledge of any process to achieve a goal or solve a problem.
Other forms of ontology, such as meta-ontology (which helps codify domain and task ontology) and knowledge representation ontology (which captures representations in knowledge represen-tation languages) can be useful to further enhance the above.
When it comes to ontologies, a picture tells a thousand words.
Figure 2 illustrates a domain ontology that represents a part of an electronic goods manufactur-ing supply chain ontology. If we consider computers to be a superset, laptops can be a subset, and multimedia laptops can be a further level of classification. These are represented by classes and subclasses in ontology.
It is important to note that computers may not be the starting point for the ontology as depicted, since they may be part of an upper ontology structure. Properties of classes will be inherited by subclasses when they can add further specialized properties. For example, the bill of materials of computers may contain some generic parts, whereas they are further specialized for laptops and then for multimedia laptops. The actual models of the laptop manufactured will be an instance
Domain Ontology
Figure 2
8 KEEP CHALLENGING November 2014
* Further specialization possible.
Process RFI/RFQ*
Plan*
Source*
Make*
Deliver
Return*
Invoice*
Metric
Measures Supports
Subclass of
Subclass of
Subc
lass
of
Subclass
of
Subclass of
Subclass of
Definition
Make Cycle Time
Instance of
Der
ives
Instan
ce of
Instance
of
Instance of Derives
Derives
Best Practice*
Process Payment*
Deliver Stocked
Receive Order*
Deliver Made-to-Order*
Calculation
Deliver Engineered-to-Order*
Return on Working Capital*
Deliver Cycle Time
Source Cycle Time
Fulfillment Cycle Time
Upper Ontology
USING ONTOLOGY TO CAPTURE SUPPLY CHAIN CODE HALOS 9
Process-Task Ontology
Figure 3
of the most specialized class. Properties such as bill of materials and product specifications may be further specialized by other subclasses.
Figure 3 illustrates a process/task ontology.6 This approach explores a part of the ontology of processes and subprocesses based on the SCOR model. The process to deliver goods is repre-sented as a class and broken down into subclasses based on the type of production approach. The delivery of stocked goods is further broken down into subclasses of the different process steps involved. The processes have metrics and follow best practices, which are represented as properties.
Metrics will have their own definitions and calculations. ”Fulfilment cycle time” and ”return on working capital” are instances of the metric class. The different subcategories of cycle times are also instances of the metric class, since they are related to the metric for overall fulfilment cycle time through a common relation. This relation will be useful to derive and calculate the master metric.
It is important to note that the above types of ontology may not necessarily sit independently of each other. The examples can potentially be part of a single master ontology definition that are interrelated with each other at some higher ontology common to both. Such a higher level of ontology can potentially be shared by multiple players of a supply chain to interconnect their systems and processes with a common depiction of information that is meaningful to them. This can lead to an amplification of their individual potential for creating higher levels of synergy.
* Further specialization possible.
10 KEEP CHALLENGING November 2014
Ontology Benefits
■ High
■ Medium
■ Low
Dat
a h
eter
og
enei
ty
Level of collaboration
Structured Data,Limited Players
Ex: Digital thermostat
Unstructured Data,Limited Players
Ex: Home entertainment system, product catalog
representation in e-commerce
Unstructured Data,Multiple Players
Ex: Smart home
Structured Data,Multiple Players
Ex: Smart grid
Applications in the High-Tech DomainIn the high-tech industry, numerous applications exist for taking an ontological approach to classify use cases by data heterogeneity and level of collaboration. Such an analysis can be performed using the 2x2 matrix proposed in Figure 4. The applications in the first, second and fourth quadrants are suitable for an ontology framework. However, maximum benefits can only be realized for applications in the first and second quadrants. Ontologies naturally lend themselves best to scenarios in which organizations must collaborate across multiple parties with unstructured data.
Example 1: Manufacturer PartneringLeading high-tech OEMs are increasingly leveraging their suppliers’ expertise in product design to improve quality and reduce costs. These collaborations – formalized through manufacturer partnership programs — are fast becoming hotbeds of innovation. However, they suffer from a flaw: Suppliers have no access to real-time product usage information and, by extension, insights into customer preferences and needs. Additional challenges include:
• Most product usage data, if recorded, lies unused with the OEM.
• There are security concerns and classification issues with sharing raw, unstructured data with suppliers.
• The amount of data is voluminous, and post-processing is extremely complex and time-con-suming.
• If post-processing takes too long, the date is often rendered meaningless as product lifecycles are getting shorter.
One possible solution is to automate the process of usage data collection and standardize relevant data available to both OEMs and suppliers. This can be done by incorporating an observation module in the product itself.7 The observation module needs to be complemented with a pre-defined observation specification to ensure that recorded data is relevant, accurate and formatted appropriately. It is critical to involve the supplier in defining the observation specification.
Framing Ontological Benefits
Figure 4
End User
Observation
ProductManufacturer
OEM
Supplier
Usage Data
Usage
Supported byProfile
Location
Has locationHas profile
Has preference
Preference
Has usage
End UserUsage metrics
Usage
Is measured by
Product Usage Observation
Observation Specifications
Records data
Observation module
Observation data analysis
Features
Identifier
Product
Sub-modules/components
Specifications Part
Manufacturer
Build parts
Supplier
Procure parts
Publish parthierarchy
Has a
Has a
Has aIs an input
Input to
Has a
Part hierarchy
Issue replacements
Build subcomponents/modules
Process & analyze usage data
Publish usage data
Build productOEM
Collect usage data
Sell product/service
Support Center
Support product/service
Is an input
USING ONTOLOGY TO CAPTURE SUPPLY CHAIN CODE HALOS 11
Furthermore, businesses should also be able to update the observation logic on-the-fly as more clarity is gained on the process. Once implemented, the pre-processed data can be shared with corresponding suppliers, who in turn, can use it to understand user preferences and incorporate them into the design.
The ontology depicted in Figure 5 provides the framework for doing just that.
Thus, the ontology can be used to “tag” data in real-time. As a result, the logged data is pre-processed and structured using high-level concepts, such as source, classification, usage, format and privacy level. Consequently, there is no need for extensive and time-consuming post-pro-cessing of raw data. Instead, relevant data can be shared with suppliers in a secure fashion, and analyzed directly and in a more efficient manner.
Example 2: Smart HomesSmart homes are a model case study for the use of ontologies in understanding Code Halos. With the advent of sensors, which are small, easily pluggable and energy-efficient, it is now possible to deploy smart-home technologies in real-world settings. Furthermore, machine-learn-ing techniques can be applied to recognize user activities from sensor-generated data. However,
An Ontology for Understanding Product Usage Data
Figure 5
Role
MoodHas role
Has mood
Has profile
Has location
Profile
Location Point
Building
GarageStoreyRoomGardenBuilding
Non-Controllable Architecture
Furniture
SensorControl
Measured by
LightingActuatorMeterPower delivery
CO2 Humidity Luminance Noise Pressure Temp Volume
Environmental parameter
HVAC System Security System
Electrical System
SystemsHome GatewayAppliances
Controllable
Has functionality
Functionality
Control Functionality
QueryFunctionality
Has command
CommandNotification
Has notification
Notification Functionality
Has screenHas memory
Device
Screen
Memory
Devices Users
Location
Building
Environment
User
12 KEEP CHALLENGING November 2014
there are challenges in the wide deployment of these techniques in ordinary homes, including the following:8
• Data heterogeneity: This can be caused by the diversity of sensors and the lack of a uniform markup language, which hinders seamless exchange, integration and reuse of data.
• Knowledge heterogeneity: This can be caused by a lack of data uniformity and formality, which prevents the sharing of defined or learned domain knowledge across different systems.
• Application heterogeneity: This can be caused by the tendency of application developers to deploy smart-home applications without leveraging the requirements for joint execution of tasks.9 Each application, meant for recognizing different human activities, works with a differ-ent architecture and operating system, which renders interoperability difficult.
These challenges can be addressed by adopting a formal, uniform ontology structure proposed by Marco Grassi (as depicted in Figure 6).10
A Formal Ontological Structure
Figure 6
Notification Command
Has command
Notification Functionality
QueryFunctionality
Control Functionality
Consumer
SupplierFunctionalityState value
Has value
State Has state
Controllable
Has functionality
Energy demand
Estimated by
Estimated Cost Appliances Home
Gateway Systems
HVAC System
Security System
Electrical System
SensorControlLightingActuatorMeterPower delivery
Renewable Nonrenewable Electric sources
Primary source Secondary source
Energy source
Has source
Produced by
Energy Supply
Energy Supplier
Has projection Has tariff
Input to
Energy tariff
Peak tariff Regular tariff
Monetary value
Estimated production
Created by
Input to
Calculated by
Has notification
Role
MoodHas role
Has mood
Has profile
Has location
Profile
Location Point
Building
GarageStoreyRoomGardenBuilding
Non-Controllable Architecture
Furniture
SensorControl
Measured by
LightingActuatorMeterPower delivery
CO2 Humidity Luminance Noise Pressure Temp Volume
Environmental parameter
HVAC System Security System
Electrical System
SystemsHome GatewayAppliances
Controllable
Has functionality
Functionality
Control Functionality
QueryFunctionality
Has command
CommandNotification
Has notification
Notification Functionality
Has screenHas memory
Device
Screen
Memory
Devices Users
Location
Building
Environment
User
USING ONTOLOGY TO CAPTURE SUPPLY CHAIN CODE HALOS 13
Using such a framework will help energy suppliers aggregate the heterogeneous information of a smart home in a semantic knowledge base. Such information can then be effectively retrieved and used in algorithms to implement intelligent control logics and task scheduling to enable, for example, efficient energy administration.
The ontology depicted in Figure 7 provides the framework for doing just that, by linking the energy supplier and the smart home consumer in a common ontology structure.11
The consumer portion of the ontology (on the right) is an extract from the smart home ontology illustrated in Figure 7. As the smart home can now access both the electricity consumption patterns and related tariffs, it can run real-time algorithms to predict and optimize energy con-sumption and costs. On the supply side, the energy producer gains visibility into the overall consumption pattern and can optimize its production and tariffs accordingly.
An Ontology for Enabling Efficient Energy Administration for Smart Homes
Figure 7
14 KEEP CHALLENGING November 2014
Identify knowledge
Define concepts and relationships
Detail concepts
Build ontology
Integrate and grow
Approach and Business OutcomesTo successfully use ontology in supply chains, organizations must take a methodical approach that involves tight collaboration among different players inside and outside the supply chain, from end-to end. We suggest an iterative approach because at every phase, new concepts or relationships may be uncovered that must be integrated into an earlier phase (see Figure 8).
The approach is bottom-up, which enables holistic coverage at the lowest levels of businesses that are collaborating across the supply chain. Developing the ontology can start as an individual activity within different business units that contribute to the overall value chain. These can then be integrated iteratively as the ontology builds across the entire organization and then finally proceeds toward a shared ontology across multiple partner organizations in a supply chain.
• Identify knowledge framework: The first phase of developing a supply chain ontology is to identify the gamut of knowledge in the functional area under focus. This typically involves identifying the boundary of all domain and process knowledge areas and the high-level con-cepts contained therein. The identification process must involve internal business experts, as well as domain specialists whenever necessary. It can also be beneficial to perform additional analysis of existing domain literature for more holistic coverage. The output of this phase is a high-level scope of the ontology.
• Define domain concepts: With the overall framework in place, the next step involves defin-ing the concepts. The concepts usually range from products, processes, services, metrics and best practices, to more specific concepts that are unique to the business. At this phase, new concepts may emerge, which usually involves re-visiting the output of the previous phase in an iterative manner. At the end of this phase, a definitive view of the concepts, which will be a part of the ontology, should be available.
• Define relationships: The third phase involves defining the relationships between concepts, which is now achievable because the domain concepts have already been defined in sufficient detail in the previous phase. These relationships will link the identified concepts using logi-cal relations, resulting in a grouping of similar concepts. This will help create hierarchies and other relations in future stages of ontology building. At the end of this phase, a definitive view of all concepts and their relationships should be available.
• Detail domain concepts: This stage covers a deep-dive analysis of the concepts and relation-ships identified in the previous phases. Here is where the output from the previous two stages is defined in detail by subject matter experts. Along with breaking down and defining indi-vidual concepts in detail, the relationships between the concepts are established at a lower level. The output of this stage will be a logical definition of the ontology in business terms.
Building a Supply Chain Ontology: Five Integrated Phases
Figure 8
USING ONTOLOGY TO CAPTURE SUPPLY CHAIN CODE HALOS 15
• Build ontology: The build phase involves using specific ontology-building tools and technol-ogy to create the ontology. Requirements of this phase include ontology definition expertise using the output of the previous phase as input, along with understanding of Web Ontology Language (OWL). Different tools, including open source, are available to build an ontology (e.g., Protégé). It is vital to choose the appropriate tool based on the intended application and the existing technology landscape with which integration may be necessary. The build output will be a completely defined ontology in the standard format, which can potentially be inte-grated with multiple systems on an as-needed basis.
• Integrate and grow: The integration and growth phase can potentially be a continuous pro-cess. Once the ontology is built, it can be shared with all supply chain partners and potentially integrated with their systems when feasible and necessary. This introduces the possibility of growing the ontology in collaboration with partners. For example, if a manufacturer has devel-oped its ontology and shares it with suppliers and distributors, the latter parties can, in turn, follow the same approach detailed above to embrace and extend the ontology. With a recur-sive approach, the entire supply chain can be covered in a common and well-defined ontology that is maintained and owned by all interested collaboration partners.
Business OutcomesWhen a common ontology is in place within and across supply chain players, it can then be enhanced and amplified through significant synergies among the players. A common ontology will provide a shareable framework for data capture, analysis, information dissemination and communication within the business and outside of it, across interlinked and inter-dependent businesses. This is akin to the emergence of a common language of communication among supply chain partners. The framework can be enhanced to align information systems and remove barriers of data exchange between applications owned and maintained by different parties, which typically run on various platforms.
The ontology that is established will be the foundation for implementing Code Halo solutions. The ontology will provide a lens to meaningfully view the multitude of data generated at various supply chain stages from Code Halo amplifiers (different devices, systems and other sources generating usable data). With all concepts and relationships already defined, it becomes much easier to develop a comprehensive set of algorithms and run analytics on this data to generate meaningful information. The ontology can further be leveraged to relate concepts that would have never been linked in a silo-based view, resulting in analytical insights that potentially amplify business value.
Looking ForwardAchieving a strategic fit among individual players has always been considered the Holy Grail of supply chain. A huge step toward achieving this goal is the emergence of a common language to classify data and interpret the information. With modern digital technology, the amount of data generated is enormous. If businesses fail to make sense of the Code Halos surrounding their supply chains — as well as their employees, their partners’ employees and the interconnected processes and smart devices that span the supply chain — they are likely to quickly fall behind the competition and ultimately be rendered irrelevant.
By using ontology effectively, businesses can create a lingua franca to take Code Halo thinking from high concept to supply chain reality.
The ontology that is established will be the foundation for implementing Code Halo solutions.
16 KEEP CHALLENGING November 2014
Footnotes1 Matthew Wall, “Big Data: Are You Ready for Blast-Off?” BBC News, March 4, 2014,
http://www.bbc.com/news/business-26383058.
2 For more on Code Halos, see our book, Code Halos: How the Digital Lives of People, Things, and Organi-zations are Changing the Rules of Business, by Malcolm Frank, Paul Roehrig and Ben Pring, John Wiley & Sons, 2014, or our white paper, “Code Rules: A Playbook for Managing at the Crossroads,” Cognizant Technology Solutions, June 2013, http://www.cognizant.com/Futureofwork/Documents/code-rules.pdf.
3 Alicia C. Böhm, Horacio P. Leone and Gabriela P. Henning, “SCONTO: A Supply Chain Ontology that Extends and Formalizes the SCOR Model,” American Insitute of Chemical Engineers, 2011, http://www3.aiche.org/proceedings/Abstract.aspx?PaperID=237484.
4 T.R. Gruber, “A Translation Approach to Portable Ontology Specifications,” Knowledge Acquisition, Vol. 5, Issue 2, June 1993, pp. 199-220, http://dl.acm.org/citation.cfm?id=173747.
5 Ali Ahmad Mansooreh Mollaghasemi and Luis Rabelo, “Ontologies for Supply Chain Management,” Industrial Engineering and Management Systems, presented at the 13th annual Industrial Engineer-ing Research Conference, Houston, May 2004, http://www2.isye.gatech.edu/people/faculty/Leon_McGinnis/8851/Sources/Ontology/Ontologies.pdf.
6 Joerg Leukel and Stefan Kirn, “A Supply Chain Management Approach to Logistics Ontologies in Informa-tion Systems,” University of Hohenheim, Stuttgart, Germany, 2008, http://link.springer.com/chapter/10.1007%2F978-3-540-79396-0_9.
7 Mathias Funk, Anne Rozinat, Ana Karla Alves de Medeiros, Piet van der Putten, Henk Corporaal and Wil van der Aalst; “Improving Product Usage Monitoring and Analysis with Semantic Concepts,” presented at the Third International United Information Systems Conference (UNISCON), April 2009, http://wwwis.win.tue.nl/~wvdaalst/publications/p532.pdf.
8 Juan Ye, Graeme Stevenson and Simon Dobson, “A Top-Level Ontology for Smart Environments,” Journal of Pervasive and Mobile Computing, Vol. 7, Issue 3, June 2011, http://www.sciencedirect.com/science/article/pii/S1574119211000277.
9 Thinagaran Perumal, Abdul Rahman Ramli, Chui Yew Leong, Shattri Mansor and Khairulmizam Samsudin, “Interoperability for Smart Home Environment Using Web Services,” International Journal of Smart Home, Vol. 2, No. 4, October 2008, http://www.sersc.org/journals/IJSH/vol2_no4_2008/1.pdf.
10 Marco Grassi, “An Ontology Framework for Smart Homes,” IEEE Workshop on Modeling and Simulation of Cyber-Physical Energy Systems, May 2013, http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5984327&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D5984327.
11 Ibid
USING ONTOLOGY TO CAPTURE SUPPLY CHAIN CODE HALOS 17
About the AuthorsShreejit Mitra is a Consultant with Cognizant Business Consulting and is a member of its Manufacturing and Logistics Practice. His areas of expertise are in the appli-cation of digital technologies in supply chain functions. He is currently involved in advising clients on ways to leverage disruptive technologies, such as social, mobility and analytics in their supply chain functions. Shreejit has an M.B.A. from XIM, Bhubaneswar, and a B-Tech in computer science and engineering. He can be reached at [email protected].
Raghu Ramamurthy is a Director within Cognizant’s High-Technology Consulting Practice. He has 14-plus years of experience in various areas of supply chain management and has worked on business transformation initiatives for clients across the U.S., Europe and APAC. His key areas of expertise include supply chain planning optimization, business process harmonization and IT roadmap devel-opment. He holds a master’s degree in management from the Indian Institute of Management Lucknow. Raghu can be reached at [email protected] | Linkedin: http://www.linkedin.com/in/RaghuRamamurthy.
Ganesh Iyer is a Manager within Cognizant Business Consulting’s Manufactur-ing and Logistics Consulting Practice. His primary areas of expertise include supply chain management and business process harmonization. He has extensive experience advising companies with supply chain planning and execution issues across manufacturing industries. Ganesh has an M.B.A. from NITIE, Mumbai. He can be reached at [email protected] | Linkedin: http://www.linkedin.com/in/ganeshiyer4 | Google+: [email protected].
Stephen Pradeep Edward has over 15-plus years of experience and has worked extensively in executing various supply chain consulting projects and programs for numerous high-technology companies, ranging from OEMs to equipment manu-facturers. His experience spans package implementation and developing custom service offerings for the high-tech segment. He can be reached at [email protected].
Aditya Dixit is a Senior Consultant at Cognizant and a member of the Cognizant Business Consulting Team in the technology practice. He has worked across diverse consulting engagements with leading high-tech and semiconductor firms. His key areas of expertise include supply chain management, trade compliance, business strategy and program management. He can be contacted at [email protected].
Ramji Mani is an AVP and Consulting Partner with Cognizant’s Manufacturing and Logistics Practice. He has over 25 years of marketing, operations and supply chain experience and is part of the consulting leadership team responsible for setting strategic direction for solutions that address client challenges and help customers align and enhance their supply chains. He can be reached at [email protected]
World Headquarters500 Frank W. Burr Blvd.Teaneck, NJ 07666 USAPhone: +1 201 801 0233
Fax: +1 201 801 0243Toll Free: +1 888 937 3277
European Headquarters1 Kingdom Street
Paddington CentralLondon W2 6BD
Phone: +44 (0) 207 297 7600Fax: +44 (0) 207 121 0102
India Operations Headquarters#5/535, Old Mahabalipuram Road
Okkiyam Pettai, ThoraipakkamChennai, 600 096 India
Phone: +91 (0) 44 4209 6000Fax: +91 (0) 44 4209 6060
© Copyright 2014, Cognizant. All rights reserved. No part of this document may be reproduced, stored in a retrieval system, transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the express written permission from Cognizant. The information contained herein is subject to change without notice. All other trademarks mentioned herein are the property of their respective owners.
About CognizantCognizant (NASDAQ: CTSH) is a leading provider of information technology, consulting, and business process outsourcing services, dedicated to helping the world’s leading companies build stronger businesses. Headquartered in Teaneck, New Jersey (U.S.), Cognizant combines a passion for client satisfaction, technology innovation, deep industry and business process expertise, and a global, collaborative workforce that embodies the future of work. With over 75 development and delivery centers worldwide and approximately 199,700 employees as of September 30, 2014, Cognizant is a member of the NASDAQ-100, the S&P 500, the Forbes Global 2000, and the Fortune 500 and is ranked among the top performing and fastest growing companies in the world. Visit us online at www.cognizant.com or follow us on Twitter: Cognizant.
