PHYSICAL-VIRTUAL CONNECTION IN UBIQUITOUS

BUSINESS PROCESSES*

Pau Giner, Manoli Albert and Vicente Pelechano

Department of Information Systems and Computation, Technical University of Valencia

46022 Valencia, Spain

Keywords: Ubiquitous Computing, Business Process, Model Driven Engineering.

Abstract: Ubiquitous Computing (Ubicomp) when applied to organizations can improve their Business Processes. An

example of improvement is the integration of real world objects in the Information System. This reduces

inconsistencies in the information and improves information acquisition rates. Automatic identification is

the key technology to achieve this paradigm shift. The present work inspects the relevance of the

identification concept for Business Processes supported by Ubicomp technologies and presents a conceptual

framework to define the elements involved in the identification process. Taking into account that Business

Processes tend to be very dynamic, it is important to have technological-independent definitions in order to

enable the evolution of systems. The presented framework is independent of the underlying technology, in

order not to be locked with one particular solution.

1 INTRODUCTION

Business Processes in organizations usually involve

real-world objects. Information Systems need

mechanisms to connect those elements at the

physical space with the information about them at

the digital space. Maintaining the connection

between both in sync is essential. Baggage loss is a

common problem in aviation industry that illustrates

the consequences of breaking this linkage. During

2006 the Association of European Airlines reported

that they had mishandled more than five million

bags (AEA, 2006). Concerned with this problem,

International Air Transport Association is

considering the usage of Radio Frequency ID

(RFID) technology

to improve logistics.

Ubiquitous Computing (Ubicomp) technologies

such as automatic identification (Auto-ID),

localization, sensor and wireless connectivity can

help to bridge physical and digital spaces. Ubicomp

represents a computation paradigm where services

are seamlessly integrated into the environment and

accessed in a natural way.

Ubiquitous Business Processes (UBPs) result from

the application of Ubicomp to support Business

Processes in organizations. The introduction of

Ubicomp in organizations is promising and

numerous benefits have been detected in economic

terms (Langheinrich, 2002), and process

improvement terms (Fleisch, 2001), (Sandner 2005),

(Strassner, 2002).

In UBPs, real world elements participating in the

process turn into intelligent interconnected objects

and the information flow between the real world and

the Information System gets automated. Without the

need for humans to act as information carriers,

information could be acquired automatically, with

more precision and at higher rates.

Identification of real-world objects is the

prerequisite for “smart” behavior (Römer, 2004).

Therefore, Auto-ID is considered the core element

for the Ubicomp enablement of systems. Auto-ID

acts as a foundation for other technologies such as

localization and moreover it represents the

automation in the linkage between real world objects

and their computational counterparts at the

Information System. This linkage constitutes the

process of identification and it is the central subject

of interest for the present work.

Different technologies can be applied to

implement Auto-ID mechanisms. Due to the

dynamic nature of Business Processes, the change in

____________________________

This work has been developed with the support of MEC under the

project SESAMO TIN2007-62894 and cofinanced by FEDER.

http://www.iata.org/stbsupportportal/rfid/

266

Giner P., Albert M. and Pelechano V. (2008).

PHYSICAL-VIRTUAL CONNECTION IN UBIQUITOUS BUSINESS PROCESSES.

In Proceedings of the Tenth International Conference on Enterprise Information Systems - ISAS, pages 266-271

DOI: 10.5220/0001697102660271

 SciTePress

requirements can have a technological impact in the

system, and the introduction –or replacement– of a

particular technology should not suppose a burden to

the evolution of the system. Defining the

identification process in an abstract way favours this

evolution, since the concepts defined become

independent of the technology used for their

implementation.

This paper is concerned with the development of

appropriate identification modelling concepts for

UBPs to treat automated and non-automated

identification in a homogeneous manner. The

contribution of this work is a conceptual framework

to reason about identification independently of the

underlying technology. In this way concepts

common to any identification technology are

defined, so the Information System can migrate

seamlessly from one technology to another.

The remainder of the paper is structured as

follows. Section 2 presents the conceptual

framework and the theoretical background in which

it relies. Section 3 defines the functionality and

components an Information system should offer to

support the presented framework. In Section 4, some

insights are given about a prototype implementation

to verify the applicability of the framework. Section

5 presents related work. Finally, Section 6 presents

some conclusions and further work.

2 THE IDENTIFICATION

CONCEPT

Identification implies to cover the gap between real

world elements and the information about them in

the Information System. Auto-ID is a particular kind

of identification process where technology is used to

automate certain operations. Different technologies

such as barcodes, infrared beacons or RFID tags

offer identification capabilities at different levels of

automation. Identification process can even be

completely manual, as it occurs in most Information

Systems nowadays. Human users are in charge of

identifying the real world elements and transmit this

information to the Information System.

Whatever physical method is used to store and

communicate the identity of an element, this

information should follow a certain codification.

Numbering schemas such as Universal Product Code

(UPC) or Electronic Product Code (EPC) are

commonly used in the consuming goods industry

and define the formats that a product identifier could

follow.

In the following sections, elements involved in

the identification process will be described. In order

to abstract technical details and formalize their

definition, concepts from the modeling area are

used.

2.1 Theoretical Basis

Our characterization of the identification idea is

based on the precise definition of the model concept

and how it is related to systems. The concept of

model is defined by Bézivin as “a simplification of a

system built with an intended goal in mind. The

model should be able to answer questions in place of

the actual system” (Bézivin, 2001). Following the

previous definition, an Information System can be

considered a model as it is a computable

simplification of a system under study used for

answering questions about it.

Favre establishes a basic taxonomy for systems

and some relation kinds among them (Favre, 2004).

Systems are classified as Physical Systems –

observable elements or phenomenons pertaining to

the physical world–, Digital Systems –residing in

computer memories and processed by computers–

and Abstract Systems –ideas and concepts that

eventually reside in human mind to be processed by

human brains–.

The most relevant relation between systems is

the RepresentationOf relation, noted as μ. This

relation establishes two roles between systems, one

system is considered the system under study and the

opposite acts as a model. Thus no system is a model

per se, being a model is a role played by a system in

relation to another system.

The concept of set and two additional

relationships are also introduced by Favre. The Set,

as defined by the set theory, is considered in the

framework as a kind of Abstract System. The

ElementOf (ε) relation is used to express the

inclusion of a system in a set. The ConformsTo (χ)

relationship is a derived relation expressing the

concept of metamodel as “a model of a modelling

language”.

These relationships presented are used to define

the concepts involved in the identification process.

The conceptual framework that includes these

concepts is presented below.

2.2 A Conceptual Framework for

Identification

The concepts involved in the identification

framework are depicted in Figure 1. The framework

defines concepts at physical, digital and abstract

spaces. An Element is part of the Physical Space. It

represents any real-world object such as a bottle in a

PHYSICAL-VIRTUAL CONNECTION IN UBIQUITOUS BUSINESS PROCESSES

267

supermarket or a package in a warehouse. The

Information System can contain information about

an element such as its price or its manufacturer. This

information is a representation of the current

element at the Digital Space. The information in the

system can be considered a projection of the

physical element in the digital space as only relevant

characteristics are present in the Information

System. An Identifier is a representation of both the

element and its digital counterpart. The Identifier is

a member of a likely infinite set of identifiers

constituting a language, a model of which is a

Codification (the Identifier is then, conformant to

this codification).

Figure 1: Conceptual framework for identification.

The common scenario is to move from the physical

space to the digital space. Identifier retrieval can be

done using a barcode reader, an RFID antenna or by

typing the element code by hand. Information in the

system should be searchable using this identifier.

And the codification used to encode the identifier

should be known.

However, the identification process involves

several transitions in both directions thru the

presented relationships. Thus the relationships

defined in the framework are bidirectional to allow

other scenarios. For example, prior to detection,

identifiers should be transferred to the physical

world.

In addition, there are no restrictions of

cardinality in the relationships presented in the

framework. An element, for example, can have

multiple identifiers. In the example, a human-

readable code (a sequence of digits printed on the

bottle label) can complement the barcode-based

identifier. Using multiple physical representations of

an identifier can help to maintain the identification

process in case of degradation of the physical

supports.

A more detailed analysis of the needed

operations that an Information System should offer

to cover the indicated transitions is presented in the

following section.

3 IMPLICATIONS FOR THE

INFORMATION SYSTEM

When the presented framework is applied to define

the identification mechanisms of an Information

System, the architecture of the identification system

should be defined. This section detects the

operations needed, which software elements should

be defined to support the framework in a modular

way and some requirements that should be fulfilled

by the Information System regarding the

identification process.

According to Kindberg, an identification-

resolution system is composed by five essential

tasks:

“Identifiers are minted, captured and converted;

and they are bound to resources. Bindings are

created and looked up. Resolution is the process of

looking up bindings from identifiers and returning

the bindings or the resources to which they refer.”

(Kindberg, 2002).

These tasks are consistent with the conceptual

framework we have previously defined. Each task

can be seen as a transition through the relations

defined in the framework, as depicted in Figure 2.

ID creation has been defined in two different steps,

the generation of the identifier following a certain

codification and the assignment to the physical

object (or minting); Identifiers can be obtained by

capturing them from the physical element or by

identifying a piece of information. Binding supposes

the creation of an association between Identifiers

and information in the system. The conversion of

identifiers produces a new identifier of a different

codification.

Figure 2: Identification tasks.

In order to keep physical and digital spaces

connected, mechanisms supporting these tasks

should be defined. Information Systems should

provide means to support them either manual or

automatically. Which components should be

considered in the design of an identification system

in conformance with the presented framework and

some remarks about their integration are given

below.

ICEIS 2008 - International Conference on Enterprise Information Systems

268

3.1 Architectural Components

The presented elements and tasks forming the

conceptual framework involve different spaces.

However, the Information System is only present at

digital space. Therefore some components at digital

space should provide the connection to physical

space, and some others should represent concepts of

abstract space. In order to support the defined tasks

in a modular way, the components defined to cover

each of these tasks are the following:

– Minters, elements capable of obtaining a

physical representation of an identifier such as a

barcode printer.

– Capturers, elements capable of obtaining an

identifier from the physical space. Barcode

readers, or RFID receivers, are some examples.

– Codifications, elements capable of producing

new identifiers in a computable form. They

contain the information limitations that the

identifier has such as a range of allowed values.

– Mediums, possible representations that a

codification can have. A barcode-based

codification can be represented by a barcode or

by a sequence of printed numbers. Barcode

readers are able to read the former while humans

are capable of reading the later.

– Transformations: define directional

transformations between two codifications.

– Information Registry: it is in charge of the

binding and information retrieval tasks.

3.2 Information System Requirements

In order to integrate the defined components and

support the identification process, the Information

System has to satisfy several requirements detailed

below.

3.2.1 Manage Codification Dependencies

Not every minter can be used to produce any

codification for any medium. There is also no

universal receiver capable of reading any identifier.

The limitation in the number of codifications a

mint, resolution or conversion system can handle

should be managed. The software component

representing any of those elements should contain

metadata indicating the supported codifications and

mediums.

Information System should allow the usage of

the elements that support a certain codification or

have a sequence of transformations to obtain this

codification.

3.2.2 Allow Complex Information Retrieval

The relation between physical elements and their

virtual counterpart is not always one to one. Physical

elements can share the same identifier or contain

multiple identifiers.

In supply stores, being impossible to tag the

product type (as it is an abstract concept), all its

members are tagged instead to represent it to retrieve

information such as prize which is not assigned in an

individual basis.

The opposite situation is also common. People

for example can be identified using email addresses,

fingerprints or organization specific identity card

numbers. Therefore, information registry should

support the possibility of having multiple identifiers,

and multiple information pieces per identifier when

information is resolved or binded.

3.2.3 Offer User-Valuable Operations

In addition to provide the basic mechanisms for

covering the essential identification tasks, more

complex operations –resulting from the composition

of the basic ones– should be considered.

Since physical elements are subjected to

degradation, Information Systems should offer

mechanisms to offer common sequences of

operations. In this way users can perform –or

program– maintenance tasks easily.

Identification regeneration is an example.

Identification regeneration consists in identifying a

piece of information and minting the identifier it has.

4 IMPLEMENTATION DETAILS

A prototype implementation was developed to prove

that the presented framework is expressive enough

to deal with different identification technologies. For

the development, RFID, fiducials, and manual

identification where chosen as identification

technologies.

The software elements supporting the different

identification tasks were componentized to promote

reusability. In this way, codifications can be used

independently of the underlying technology, as far

as the types supported by the technology form a

superset of the required for that codification. For

example, Serial Shipping Container Code (SSCC) is

a eighteen digit codification that can be used in

EPC-based RFID systems as well as in barcodes.

However, the standard fiducial set contains only 90

symbols, so it cannot support this codification.

Web Services technology has been chosen to

form the architecture of the system as it is transport

PHYSICAL-VIRTUAL CONNECTION IN UBIQUITOUS BUSINESS PROCESSES

269

independent and enables the definition of distributed

systems following a Service Oriented Architecture.

This allows the integration of identification systems

from different vendors.

The fiducial based identification subsystem is

based on reacTIVision

. Fiducials –specially

designed markers recognizable in a real time video

stream– are used to identify elements (Bencina,

2005).

Figure 3: A ﬁducial and its left heavy depth sequence

(Bencina, 2005).

The usage of fiducials results in an intuitive and

affordable identification solution. The complete

Auto-ID system is composed by a USB video

camera as a capturer, a common printer as a minter

and the reacTIVision framework to handle the

encoding and decoding tasks. The codification

consists in a module that generates numerical values

from 0 to 90. The printer supports two mediums for

this codification, one corresponding to the fiducial

image corresponding to the code in the standard

fiducial set, and another just by printing the number.

These components of the infrastructure have been

correspondingly wrapped in order to be accessed as

Web Services.

Figure 4: Accada Reader Simulator.

Identification subsystem supporting RFID is based

on Accada

, an open source RFID prototyping

platform. The Accada platform deals with readers,

filters and RFID data aggregation following

EPCglobal standards, the predominant

standardization effort of the RFID community. The

usage of such middleware eases the development

and reduces the technological heterogeneity. In

Figure 4 the simulation tool used to validate the case

study is shown.

Web services are also used as a wrapper for the user

interaction when acting as an identification system.

In this last scenario a web application was used to

allow the user to introduce a numeric code and to

obtain system information.

5 RELATED WORK

The idea of automating the linkage between real and

virtual elements is nothing new. Research has

considered the application of printed markers (Ishii,

1997), (Ljungstrand, 2000), (Rekimoto, 1995), RFID

technology (Floerkemeier, 2005), (Römer, 2004),

(Want, 1999) or infrared beacons (

Kindberg et al.,

2002) as a way to enforce this linkage.

Some frameworks for the development of

Ubicomp applications such as One.world (Grimm,

2002), Interactive Workspaces (Johanson, 2002),

Cooltown (Kindberg et al., 2000) or Gaia OS

(Roman, 2002) have been developed to abstract this

technological heterogeneity. Auto-ID specific

middleware has been also developed such as

Accada, SAP’s Auto-ID infrastructure (Bornhövd,

2004), Java RFID Api (Sun 2006) or OAT C4

Architecture (OATSystems 2006). These

infrastructures provide useful programming

primitives to filter and aggregate information from

Auto-ID devices and transmit application-relevant

events. The APIs provided by these frameworks

supply relevant information at application level

allowing developers to build their systems on top on

them.

Providing constructs to represent the needed

mechanisms in a technological agnostic way is what

this work tries to address, so these infrastructure

elements are complementary to the objectives of the

present work.

6 CONCLUSIONS

The present work introduces a conceptual

framework to define in a precise way the elements

involved in the identification process and the

relationships among them. This is useful to

determine the operations and components that an

Information System supporting identification should

provide independently of the used technology.

In the present work the identifier is not considered to

have a meaning. In some numeric schemas, it has

several parts to indicate company, location reference

and other information and the like. With the present

implementation, the elements in charge of

codification should format the identifier correctly.

_____________________________________

http://www.iua.upf.es/mtg/reacTable/?software

http://www.accada.org

ICEIS 2008 - International Conference on Enterprise Information Systems

270

How these elements are linked with the specific

information constitutes further work.

The capabilities to store information that

technologies such as RFID have, has not been

considered in the present work. Considering the

possibility of distributed information across different

Information systems – or storing devices such as the

RFID tag– requires the definition of mechanisms to

maintain information consistency.

REFERENCES

Association of European Airlines (AEA) (2006) Consumer

report.

Bencina, R.; Kaltenbrunner, M.; Jorda, S. (2005),

Improved Topological Fiducial Tracking in the

reacTIVision System, Computer Vision and Pattern

Recognition, 2005 IEEE Computer Society

Conference on , vol.3, pp. 99-99, 20-26.

Bézivin, J., Gerbé, O. (2001), Towards a Precise

Definition of the OMG/MDA Framework. In:

Proceedings of ASE’01

Bornhövd, C., Lin, T., Haller, S., Schaper, J. (2004),

Integrating Automatic Data Acquisition with Business

Processes - Experiences with SAP’s Auto-ID

Infrastructure. In Proceedings of the 30st international

conference on very large data bases (VLDB), pages

1182–1188, Toronto, Canada.

Favre, J.M. (2004), Foundations of Model (driven)

(Reverse) Engineering: Models - Episode I: Stories of

the Fidus Papyrus and of the Solarus. In: post-

proceedings of Dagsthul Seminar on Model Driven

Reverse Engineering.

Fleisch, E. (2001). Bussiness Perspectives on Ubiquitous

Computing. M-Lab Working Paper No. 4. University

of St. Gallen, Switzerland, 2001.

Floerkemeier C., Lampe, M. (2005), RFID middleware

design – addressing application requirements and

RFID constraints. In Proceedings of SOC’2005, pages

219–224, Grenoble, France.

Grimm, R., (2002), System support for pervasive

applications, Ph.D. Thesis, University of Washington,

Department of Computer Science and Engineering.

Ishii, H., Ullmer, B. (1997), Tangible Bits: Towards

Seamless Interfaces between People, Bits and Atoms.

In: CHI ’97: Proc. of the SIGCHI conference on

Human factors in computing systems, ACM Press

234–241

Johanson, B., Fox, A., Winograd, T., (2002), The

interactive workspaces project: Experiences with

ubiquitous computing rooms, IEEE Pervasive

Computing 1(2) 71–78.

Kindberg, T. (2002), Implementing physical hyperlinks

using ubiquitous identifier resolution, in 'WWW '02:

Proceedings of the 11th international conference on

World Wide Web', ACM Press, New York, NY, USA,

pp. 191—199

Kindberg, T., Barton, J., Morgan, J., Becker, G., Caswell,

D., Debaty, P., Gopal, G., Frid, M., Krishnan, V.,

Morris, H., Schettino, J., Serra, B., Spasojevic, M.

(2000), People, Places, Things: Web Presence for the

Real World. Mob. Netw. Appl. 7(5) (2002) 365–376

Langheinrich, M., Coroama, V., Bohn, J., Rohs, M.

(2002), As we may live - Real world implications of

ubiquitous computing. Technical Report, Institute of

Information Systems, Swiss Federal Institute of

Technology, Zurich, Switzerland.

Ljungstrand, P., Redström, J., Holmquist, L.E. (2000),

WebStickers: Using Physical Tokens to Access,

Manage and Share Bookmarks to the Web. In: Proc. of

DARE 2000 on Designing augmented reality

environments, ACM Press 23–31

OATsystems (2006). OAT C4 Architecture.

Rekimoto, J., Nagao, K. (1995): The World through the

Computer: Computer Augmented Interaction with

Real World Environments. In: UIST ’95: Proc. of the

8th annual ACM symposium on User interface and

software technology, ACM Press 29–36

Roman, M., Hess, C., Campbell, R. (2002), Gaia: An OO

middleware infrastructure for ubiquitous computing

environments. In: ECOOP Workshop on Object-

Orientation and Operating Systems (ECOOP-

OOOSWS). Malaga, Spain.

Römer, K., Schoch, T., Mattern, F., Dübendorfer, T.

(2004): Smart Identification Frameworks for

Ubiquitous Computing Applications. Wireless

Networks 10(6) 689–700.

Sandner, U., Leimeister, J. M., Krcmar, H. (2005),

Business Potentials of Ubiquitous Computing. In:

Proceedings of the Falk Symposium. No. 146 Gut-

Liver Interactions: Basic and Clinical Concepts.

Innsbruck, Austria.

Strassner, M., Schoch, T. (2002), Today’s Impact of

Ubiquitous Computing on Business Processes. In:

Pervasice Computing. First Int. Conf. Pervasive

Computing 2002, Zurich, Switzerland.

Sun Microsystems (2006). Java System RFID Software

3.0 Developer Guide. www.sun.com.

Want, R., Fishkin, K. P., Gujar, A., Harrison, B.L. (1999),

Bridging Physical and Virtual Worlds with Electronic

Tags. In: CHI ’99: Proc. of the SIGCHI conference on

Human Factors in Computing Systems, Pittsburgh,

PA, USA 370–377

PHYSICAL-VIRTUAL CONNECTION IN UBIQUITOUS BUSINESS PROCESSES

271