Individual Service Clearing as a Business Service: A Capable Reference
Solution for B2B Mobility Marketplaces
Michael Strasser
1
, Nico Weiner
1
and Sahin Albayrak
2
1
Bosch Software Innovations GmbH, Sch
¨
oneberger Ufer 89, 10785, Berlin, Germany
2
DAI Laboratory, Technical University of Berlin, Ernst-Reuter-Platz 7, 10587, Berlin, Germany
Keywords:
Individual Service Clearing, Service Marketplaces, Service Platforms, Service Interface Design, Smart City
Mobility Services.
Abstract:
The paper presents an approach for individual and outsourced service clearing within Business-to-Business
marketplaces for mobility services in a Software as a Service fashion. The lack of service clearing possibili-
ties within today’s marketplaces solutions for mobility services have been confirmed by interviews experts who
highlight the need for clearing. To enable service clearing, appropriate interfaces which enable access upon
data are required. Current solutions lack those interfaces and thus service clearing to charge service transac-
tions is not possible. The paper discusses interfaces for outsourced service clearing according to their design,
data and role dependencies. A reference solution is implemented to demonstrate the feasibility of the inter-
faces and the overall clearing approach. A clearing algorithm has been developed to validate the interfaces’
reliability and correctness. Our presented clearing approach enables marketplace participants to outsource the
transaction clearing to a provider which offers clearing capabilities. The work on hand contributes to interface
and protocol standardization in respect to service clearing and marketplace interconnectivity.
1 PROBLEM DESCRIPTION
The number of service consumers and providers for
mobility services (e.g. car sharing, charging, park-
ing, routing) increases constantly. It is anticipated that
this trend will probably result in a higher number of
service offerings, demands and transactions (Thitima-
jshima et al., 2015). As a logical consequence, more
business relationships will be established via elec-
tronic service marketplaces. These marketplaces rep-
resent the environment in which participants can ac-
complish service trades. These marketplaces are pure
B2B oriented and do not deal with end-customers.
End-customers are registered with service consumers.
The increasing demand for convenient mobility will
subsequently increases the usage mobility services.
That circumstance leads to more transactions which
are processed by a service operator and probably
routed by a marketplace. One of the service operator’s
duties is to have capabilities to uniquely assign a sin-
gle transaction to one particular contract (Pfeiffer and
Bach, 2014). This is necessary to assure that a ser-
vice consumption (transaction) is cleared (charged)
according to the conditions defined in the business
contract.
It is likely that the growth of service consump-
tions results in a higher effort for service clearing
(transaction billing). Therefore it is feasible to as-
sume that certain service operator want to outsource
the complete service clearing process to a third party.
This party has to have comprehensive knowledge and
approved and standardized processes in the service
clearing domain. For individual service clearing, a re-
spective clearing operator has to register with a mar-
ketplace and offer clearing services to other partici-
pants as software as a service (SaaS). Such a clearing
operator is called service consumption clearing op-
erator (SCCO). A service operator, for instance for
charging stations, who want to outsource the service
clearing process can contract a SCCO’s clearing ser-
vice. Once they are business partners, the charg-
ing station operator can assign the SCCO’s clearing-
service to all his services which he wants to be cleared
(charged) by the SCCO. Therefore, a service market-
place for mobility services has to provide mechanisms
to ensure accurate data access on internally stored
data. Interviewed experts
1
point out that individual
1
The interviewees’ affiliation are presented in the ac-
knowledgment section.
Strasser, M., Weiner, N. and Albayrak, S.
Individual Service Clearing as a Business Service: A Capable Reference Solution for B2B Mobility Marketplaces.
In Proceedings of the 6th International Conference on Cloud Computing and Services Science (CLOSER 2016) - Volume 2, pages 27-38
ISBN: 978-989-758-182-3
Copyright
c
2016 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
27
service clearing does not exist in electronic market-
places for mobility services but is required. Individual
in this context means that a marketplace participant
can do the clearing on his own or contract and entitle
a professional SCCO. However, due to missing Ap-
plication Programming Interfaces (API)s, outsourced
service clearing is one of many use cases which can-
not be accomplished because access to data stored
within the marketplace is not possible. As discussed
by Gubbi et al., a service infrastructure has to provide
appropriate APIs which can be used to enable flexible
billing (Gubbi et al., 2013).
The rest of the paper is organized as follows. Sec-
tion 2 presents the current state of the art in B2B ser-
vice clearing in marketplaces for mobility services.
Section 3 introduces the interfaces which are deemed
as necessary to enable communication with the mar-
ketplace. Section 4 discusses the elaborated approach
for individual service clearing. The evaluation of the
interfaces and the elaborated and prototypically im-
plemented service clearing algorithm are presented in
Section 5. The paper ends with a conclusion in Sec-
tion 6 and a short outlook on further research in Sec-
tion 7.
2 CURRENT STATE OF THE ART
Research projects
2
explore possibilities about how to
provide services for the mobility domain, like charg-
ing services for electric vehicles, sharing services for
bike or car as well as services which help to find free
parking lots. A system which provides access to such
services is called for example e-hub (Grieger, 2003),
service provisioning platform, (
˚
A gerfalk et al., 2006),
Value Added Service Supplier (Legner, 2007) inter-
mediary (Heinrich et al., 2011), Platform (Buchinger
et al., 2013) or electronic marketplace, trading com-
munities, trading exchanges (Turban et al., 2015).
Companies
3
also provide Information and Communi-
cation Technology (ICT) based solutions over which
mobility service operators can offer and mobility ser-
vice consumers can access mobility services. These
mobility services enable end-customers of service
consumers to experience the capabilities of future
mobility. An electronic marketplace is the environ-
ment for service operators (offer services) and service
consumers (consume services) to trade services like
2
For example: Green eMotion (Green-eMotion, 2015),
CROME (CROME, 2014), Olympus (Olympus, 2015),
Streetlife (Streetlife, 2015)
3
For example: Hubject (Hubject, 2015), Parku (Parku,
2015), Multicity (Multicity, 2015), Smartlab (smartlab In-
novationsgesellschaft mbh, )
goods (Ghenniwa et al., 2005). An electronic market-
place matches sellers and buyer and encourage them
for service trades by providing appropriate trading ca-
pabilities (Turban et al., 2015).
Current marketplaces for electric mobility ser-
vices lack the possibility to enable data access on
internally processed and data from outside. This is
a limitation and prevents further use cases and the
realization of Critical Success Factors (CSF). Inter-
viewed experts have emphasized that it is mandatory
to access transaction data, contract data, participant
information or service descriptions via APIs. By the
current date, neither an appropriate protocol nor re-
spective interface standards are available to guide the
development of respective interfaces regarding com-
prehensive communication with the outside world.
Without comprehensive data access, internal data is
locked inside a marketplaces and cannot be processed
further by any participant expect the marketplace it-
self.
Costs accrue for being a member of a marketplace
(depends on a marketplaces’ revenue model (John-
son, 2013)) but also for using a service provided via
a marketplace (interviewed experts). An identified
business case (also indicated by interviewed experts
and literature) is business to business (B2B) clearing
also known as service charging. Clearing is consid-
ered as a CSF of electronic marketplaces (Balocco
et al., 2010; Johnson, 2013). The experts point out
that service provisioning marketplaces have currently
no elaborated clearing approach or completely lack
the possibility to enable service clearing within their
system. Pfeiffer and Bach point out that a business
model for clearing does currently not exists and that
”the billing process is more cost-intensive than the
overall billing amount” (Pfeiffer and Bach, 2014).
These are the reasons why it is currently not op-
erated within marketplaces or platforms. However,
clearing mechanisms have to be in place as trans-
actions are not free of charge (Rust and Zahorik,
1993; Buyya et al., 2008; Johnson, 2013; Thitima-
jshima et al., 2015). Even though no appropriate busi-
ness models for transaction clearing exists, the rise
of new demands or the development of new capabil-
ities might enable cross selling or combining of ca-
pabilities which justifies certain clearing models. A
clearing service is deemed as a value added service in
within electronic commerce conducted via a market-
place (Turban et al., 2015).
Marketplaces for electric charging infrastructure
like Green eMotion and Smartlab’s e-clearing.net per-
form do B2B service clearing for their participants
right within their systems. There is no choice for
their participants to appoint another clearing operator.
CLOSER 2016 - 6th International Conference on Cloud Computing and Services Science
28
CROME or Hubject do not support service clearing
inside their platform and leave it to their participants
who have to agree on a common clearing method.
The limited service clearing mechanism is a disad-
vantage of all of today’s platforms and marketplaces
for mobility services. Current solutions do not pro-
vide appropriate interfaces that enable access to inter-
nally stored data or do not even store data accordingly
(interviewed experts). This cripples the possibility
for individual service clearing where participants out-
source the clearing process and appoint an individual
clearing service operator which meets their require-
ments. The state of the art in transaction clearing
in marketplaces for mobility service is in summary
that either the marketplace itself seizes service clear-
ing or participants have to do it somehow outside the
marketplace. Marketplaces like Amazon or iTunes as
well as eBay and other known marketplaces or trad-
ing platforms offer a handful payment options to their
participants. Individual service clearing via SCCOs
and internal B2B transaction clearing is not foreseen
in their solutions.
The paper on hand is a comprehensive expansion
of (Strasser et al., 2016). Its contributions are pre-
sented in List 1 below.
A comprehensive outline of the problem domain
and the current state of the art
An extended overview of required interfaces for
future marketplaces along with an simplified ar-
chitecture
An expanded presentation of the defined clearing
process
A deeper discussion of the elaborated clearing in-
terfaces and their implementation
An extensive presentation of the developed ser-
vice clearing protocol and its parameters
An open demonstration of the elaborated clearing
algorithm
An performance evaluation of the clearing algo-
rithm
List 1: Contribution of the paper.
3 MARKETPLACE INTERFACES
Marketplaces have to provide appropriate APIs to en-
able access on internal data which is fundamental
for outsourced and individual service clearing. A
comprehensive set of marketplace interfaces enables
the accomplishment of various additional use cases,
for example interconnectivity between service mar-
ketplaces (Strasser and Albayrak, 2015; Strasser and
Albayrak, 2016) or Service Level Agreement (SLA)
and condition monitoring. The marketplace devel-
oped within this work consists of Service Oriented
Architecture (SOA). The marketplace’s communica-
tion interfaces use the Simple Object Access Protocol
(SOAP). The interfaces are developed in a role-based
manner instead of a data-driven manner. The in-
terfaces are furthermore differentiated between pub-
lic and private. Context, service or participant rele-
vant data can only be accessed via private interfaces.
This kind of data might be classified as confidential
and only accessible for registered marketplace partic-
ipants. General information about a marketplace and
its participants, services, supported domains or oper-
ation area can be retrieved via public interfaces which
do not require a marketplace participation. List 2
presents interfaces which have been identified to be
of public interest. Thus they are accessible without
any marketplace registration. Registered participants
can use those interfaces too.
Public interfaces:
GetMarketplaceInformation
GetParticipantOverview
GetServiceOverview
GetMarketplaceConditions
SetRegisterParticipant
List 2: Public marketplace interfaces.
List 3 provides an abstract of private interfaces
which have been identified to be of private interest.
They provide access to confidential and participant
dependent information and thus can only be accessed
by registered participants. These interfaces are as-
sumed as necessary to enable the communication with
other service marketplaces as well as to enable indi-
vidual and outsourced service clearing. Of course, re-
spective processes have to be implemented too. The
interfaces for the outsourced service-clearing are em-
phasized in bold. The interfaces for the marketplace
interconnectivity are emphasized in italic. The inter-
faces are named according their functionality and pur-
pose.
Set and Get indicate whether information can be
retrieved (pull) from or send (push) to a marketplace.
The interfaces’ grouping towards public or private is
not final. It might change once more use cases be-
come reasonable. Interfaces which are listed in both
lists provide more information if they are applied in
private mode. Due to security mechanisms, a API re-
quester is validated if it is a registered marketplace
participant. This is done by a certificate. Then it is
Individual Service Clearing as a Business Service: A Capable Reference Solution for B2B Mobility Marketplaces
29
Private interfaces:
GetParticipantOverview
GetServiceOverview
GetParticipantInformation
GetServiceInformation
SetCreateServiceOfferQuotation
SetCreateServiceSearchQuotation
GetServiceOfServiceType
GetTransactionData
GetClearingRelatedContracts
GetContractData
SetUploadMarketplaceData
SetUploadParticipantData
SetCloseContract
SetUploadAllAvailableServices
GetEndCustomerAuthorization
GetMarketplaceConditions
GetSearchServiceQuotation
SetAcceptServiceQuotationCondition
List 3: Private marketplace interfaces.
checked what role the requester has to decide what
data can be accessed.
The interfaces are role-base driven rather than
data driven. That enables re-using and re-purposing
of interfaces while being able to limit the data ac-
cess accordingly. For security reasons, private and
public interfaces are separate web-services. This is
done to relieve each interface and because the in-
terfaces can have different service level agreements
Interfaces layer
Public:
GetParticipantOverview
GetServiceOverview
SetRegisterParticipant
Private:
GetParticipantOverview
GetServiceOverview
GetContractData
Security layer
(user credentials, role,
certificate)
Service
Quotations
Service
Consumer
Marketplace
Service
Operator
no security
Marketplace
Contracts
Transactions
Internal data layer
Figure 1: Interfaces to access internal marketplace data.
(SLA)s. Each web-service consists of various opera-
tions which represent one of the above identified in-
terfaces. Extensible markup language (XML) schema
validation is used to validate the input data. This en-
sures data type integrity and avoids malfunction. The
private web-service validates user-credentials and cer-
tificates. Figure 1 presents a simplified overview
about how marketplace data is accessed from exter-
nal.
Security mechanisms have to be in place to val-
idate requests and to determine which data the re-
quester is allowed to access. Even though the security
layer can also be applied for public interfaces, it is as-
sumed that security checks are costly and thus should
only be conducted when necessary.
4 ACCESSING AND PROCESSING
INTERNAL MARKETPLACE
DATA
The proposed marketplace interfaces facilitate the re-
alization of various use cases. Request routing, per-
formance and efficiency monitoring, Service Level
Agreement (SLA) validation or any other data pro-
cessing or measuring use cases can be developed.
However, the interfaces also enable individual service
clearing for B2B service transactions. Clearing ser-
vices require access at internal marketplace data from
outside. The elaborated marketplace provides those
interfaces and processes which are necessary to ac-
cess all required data to create an invoice for service
consumption. Service-clearing particularly requires
access to contract and transaction data.
4.1 Service Consumption Data
The following mobility use case exposes how trans-
action data incurs within a mobility service market-
place. A service operator (SO) has parking infras-
tructure. SO is registered with a service marketplace
(MP) and offers a service via the MP which enables
to access his infrastructure. The service’s functional-
ities are for example i) to open a barrier arm, ii) to
guide a driver to an empty parking spot or iii) to de-
termine the duration of the parking. Service consumer
(SC) is also registered with MP. SC searches for a ser-
vice with capabilities to access parking infrastructure.
Therefore, SC i) contacts SO, ii) negotiates the ser-
vice’s conditions and iii) settles a contract with SO.
A digital representation of the paper contract is stored
inside MP. The end-customer (EC) of SC cannot, at
first, access the parking infrastructure of SO because
CLOSER 2016 - 6th International Conference on Cloud Computing and Services Science
30
SO does not know EC. For SO is EC a stranger and
thus forwards the request to MP. MP checks all of
SO’s contracts and forwards the service request to all
of SO’s business partners. SC retrieves the service re-
quest and recognizes EC. SC responses accordingly
and MP forwards the response to SO. SO opens his
parking infrastructure for EC because SC has acted
as EC’s guarantor. As soon EC leaves the parking
area, SO sends a park detail record, which contains
information about the parking process, to MP which
forwards the record to SC.
4.2 Individual Service Clearing - An
Outsourced Invoice Approach
In the current context, service clearing is done in
respect to B2B service consumption among market-
place business partners (service operator and con-
sumer). Outsourcing the service clearing process to
a third party calls Luttge outsourced invoice (Luttge,
2001). This approach is, according to the interviewed
experts, not implemented by any mobility service
marketplace. However, the experts assume it as a fun-
damental functionality and according to Turban et al.,
a trading platform should provide mechanisms to ar-
range payment (Turban et al., 2015). Even though
Vidal et al. (Vidal et al., 2011) describe in detail what
kind of data has to be in place to clear an electric vehi-
cle charging process via a service marketplaces, they
miss the chance to implement and present a working
solution.
To outsource the service clearing process to a
SCCO a service operator must comply to the follow-
ing process: A SCCO-A offers a clearing service via a
marketplace (MP). Another service operator (SO-B),
who offers parking infrastructure services, closes a
contract with SCCO-A. Then, SO-B assigns the con-
tracted clearing service of SCCO-A to his service XY.
This entitles SCCO-A to access those contracts and
transactions which relate to SO-B’s XY service. The
service description of XY should particularly empha-
size that a third party (SCCO-A) has been entitled to
accesses all accrued transactions and the contracts re-
lated to service XY. A service consumer (SC) who
signs a contract with SO-B for service XY agrees on
the service’s conditions and thus accepts that SCCO-
A accesses all clearing relevant data. This agreement
is important because the transaction data represents a
kind of an end-customer’s mobility profile. Thus SC
has to inform his end-customers about the third party
access. SCCO-A gathers the necessary data accord-
ing to the defined clearing schedule and calculates the
price which SC has to pay SO-B. Due to the individ-
ually, SO-B can appoint another SCCO to his other
services but also can do the clearing on his own with-
out third party support. This functionality is not in
place by current available solutions but is prototypi-
cally implemented within this research.
The role-heritage diagram in Figure 2 presents the
roles used in the use-case diagram presented in Fig-
ure 3. The role Clearing Service Provider is not a
individual role in particular but represents a possible
character of a service provider. The generic use-case
diagram illustrates the roles, systems and processes
which are part of a individual and outsourced clearing
process. The elaborated Use-Case diagram is neither
an extension nor a modification of (Pfeiffer and Bach,
2014) e-roaming clearinghouse. It presents a succeed-
ing step in which e-roaming is actually charged ac-
cording to negotiated conditions written in a contract
during, what (Pfeiffer and Bach, 2014) calls, settle-
ment.
Participant
Register
Create Quotation
Establish
Relationship
Service
Provider
Service
Consumer
Clearing
Provider
Figure 2: Overview role inheritance.
To achieve individual and outsourced service
clearing various logical and functional challenges
have to be approached. An abstract of identified chal-
lenges is shown in List 4. The list does not claim
to be complete. Once our proposed solution is im-
plemented in commercial solutions it is believed that
further challenges will be identified. The challenges
furthermore depend on the design of the service clear-
ing as well as on the implemented contracting mech-
anism.
Legal regulations are not part of the research’s
scope. The interviewed experts confirm the com-
plexity of service clearing but unfortunately do not
propose a suitable solution. No standardized pro-
tocol or data set exists on which data exchange via
marketplace APIs can be build on. The relevant do-
main protocols have capabilities to support the mar-
ketplace in its operation only. Marketplace partici-
pants are able provide to information to the platform
but cannot retrieve necessary information about con-
tracts and transactions or other data useful for future
business cases. Access on data is only possible for
data which has been provided by the participants be-
fore. No guidelines are available which emphasize the
required steps to enable data access from outside and,
Individual Service Clearing as a Business Service: A Capable Reference Solution for B2B Mobility Marketplaces
31
<<subsystem>>
Clearing
Provider
Backend
Collect data
Create invoice
Send invoice
<<subsystem>>
Service Provider
Backend
Connect
Infrastructure
Grand Access
Pay Clearing
Provider
Check Payment
<<subsystem>>
Service
Consumer
Backend
Register End
Customer
Authorize End
Customer
Charge End
Customer
Pay Consumption
Invoice
Service
Provider
<<Service Marketplace>>
Service
Consumer
Register
Establish Business
Relationship
Create Service
Quotation
<<requires>>
<<requires>>
Set particular
Clearing Provider
<<extends>>
Offer
<<extends>>
<<extends>>
<<requires>>
Transaction
Data
<<requires>>
<<requires>>
End
Customer
requires
Search
Clearing
Service
Provider
Identify
Pay Consumption
Charge Service Consumer
for Service Consumption
Consume service
Participant
Contract Data
Figure 3: Use-Case diagram for individual service clearing.
Authorize access at other participants’ data
Control data access and check if it is valid
Avoid multiple data access by multiple SCCOs
Identify which data is allowed to be accessed
Notify service consumers about SCCOs data ac-
cess
Define process if a service’s SCCO has changed
Validate how often data is accessed
Determine processes and tasks to close a clearing
contract
Avoid that a SCCO outsources his clearing duty
to other SCCOs
Evaluate the robustness of the interfaces
Evaluate whether interfaces can be used for cross-
marketplace service clearing
List 4: Challenges according to service clearing.
even more important, no research was found which
identified what interface functionalities are required
by mobility marketplace participants. Therefore, this
work provides insights into the required steps, tasks
and data for doing so while identifying the need for
individual and outsourced service clearing.
4.3 Service Clearing Interfaces
Table 1 presents the elaborated interfaces required for
outsourced and individual service clearing. These in-
terfaces are used by a SCCO to access clearing rel-
evant data. The clearing functionality has been re-
alized with several light wise interfaces to achieve a
better performance, to reduce maintenance work and
to enable interface re-usability.
It is pointed out that marketplace interfaces should
not, if applicable, serve one particular use case only.
Use case relevant interfaces process and return data
for a particular use case, thus they are probably not
applicable for other use cases. The developed market-
place and its interfaces are use case independent and
role-based driven. Thus participants and there roles
are differentiated and the data returned respectively.
The interfaces in Table 1 are examples of role-based
driven interfaces because they return data according
to the provided certificate and identifiers. The first
can only be used, at the moment, by a SCCO while
the remaining two can be used by all registered mar-
ketplace participants for informative reasons. A role
validation is necessary to decide which data has to be
CLOSER 2016 - 6th International Conference on Cloud Computing and Services Science
32
Table 1: Interfaces for service-clearing.
Interface Name Description
GetClearingRel -
atedContracts
Returns a list of contracts in
which the requester is set as
SCCO
GetContractData Returns contract details like
participant contact data and
usage price plan
GetTransaction -
Data
Returns transaction data re-
lated to the contract and the
SCCO’s client
delivered to the requester.
Due to the interfaces’ use case independence im-
plementation, they can be used individually or in
combination. Individual usage would be, for exam-
ple, to check transactions or to check when a specific
contract expires. Combined usage is about using an
interface’s output for another interface’s input. This
is done for service clearing and presented in Figure 4.
At first, a SCCO checks his contracts to identify his
clients by using the GetClearingRelatedContracts in-
terface. The contracts define the services for which
the SCCO is entitled to do service clearing. Once
SCCO knows which services are under his clearing
responsibility, SCCO uses the GetContractData inter-
face to retrieve the contracts which have been signed
between his clients and their business partners. With
those contracts SCCO knows the service consumption
conditions (price plan) and other contact data. The
GetTransactionData interface is used to retrieve all
transaction data related to a specific service. Once
all data is gathered a SCCO computes the costs of
all retrieved transactions according to the contract and
the specified price plan. A potential capability might
be that SCCO sends the final invoice to his clients’
business partners on behalf of his client. As previ-
ously outlined, a service operator is able to entitle
different clearing operators to do the service clear-
ing for different services but only one SCCO per ser-
vice at the same time. From a SCCO point of view
it is relevant which interface is used first. The pre-
sented input and output parameters in Figure 4 are a
summary of the most important parameters. Each in-
terface requires a minimal set of input information.
To avoid the accidentally disclose of data, the inter-
face returns as little information as necessary in ac-
cordance to the requesters role and the role’s scope.
This procedure contributes to the assurance of privacy
and security. The marketplaces’ interfaces are devel-
oped to actively contribute to protocol standardization
for marketplace in the mobility domain and therefore
the protocols parameters are presented. The lessons
learned while develop the presented processes and in-
terfaces can be used for the implementation of addi-
tional interfaces necessary for the use case introduced
in the beginning. This is also a reason why indepen-
dent but role-based driven interfaces are required.
Figure 5 presents a Business Process Diagram
(BPD) which demonstrates the access via the Get-
TransactionData interface with all validations. If the
input parameters pass all validations then the transac-
tion data is returned.
4.4 Interface Implementation
The interfaces have been implemented using inubit, a
Java and XML based Business Process Modeling ap-
plication. The process engine runs on Apache Tom-
cat. The AccountId is the most general identifier and
is used to identify a requester. The ContractId is in-
dispensable to determine a contract and subsequently
retrieve the contract details. The prototypical imple-
mentation of the introduced clearing interfaces has
proven that these two identifiers are sufficient to deter-
mine all account, contract and transaction data. Spe-
cial roles like SCCO have separate identifiers to avoid
confusion when using the interfaces. The GetClear-
ingRelatedContracts is an interface that specifically
requires a SCCOID to uniquely identify a clearing
operator rather than a general AccountId. The SC-
COID replaces the AccountId. All interfaces check
whether an AccountId fits to the given ContractId or
not. This is done to determine a requester’s access
scope. For example, using the GetTransactionData
interface with the same ContractId but different Ac-
countIds might returns with i) a different data set or
ii) with a subset of the former return data or iii) with
the very same data. The reason is that the given Ac-
countId is a) set in the contract as business partner
or b) is set as a clearing operator which is allowed to
access this particular data. The interfaces need to be
designed to detect and differentiate all possible sce-
narios and respond accordingly. This is also presented
in Figure 5.
Figure 6 on the next page shows the GetCon-
tractData response. It contains the ContractId and
all contract details. The contract details depict the
business partners. The service operator is shown in
ProviderAddress and the respective service consumer
in ConsumerAddress. The service operator entitled
the SCCO to do the service-clearing on its behalf. The
price conditions are necessary to charge each transac-
tion individually in accordance to the defined usage
price plan. A operator of a service can define a ba-
sic price for each service transaction. However, it is
also possible that special time slots are defined. The
Individual Service Clearing as a Business Service: A Capable Reference Solution for B2B Mobility Marketplaces
33
GetTransactionData
Input:
Account ID
Clearing Contract ID
From Date
Till Date
Output:
List of transactions
with details
Calculate Costs
(SCCO Backend)
Input:
Contract Details
List of transaction with details
Output:
Invoice for service-consumption
which will be send to service
consumers on behalf of service
operator
GetClearingRelated
Contracts
Input:
Account ID (SCCO)
Output:
Contracts where
SCCO does service
clearing
GetContractData
Input:
Account ID (SCCO)
Clearing Contract ID
Output:
Contract Details
Figure 4: Interface sequence for service clearing.
Figure 5: Process to access own or third party transaction
data.
prototypically implementation has shown that prices
of such time slots can differ. During the prototyping
it was detected that the time slots require a validation
as they should not overlap each other. Furthermore
would it be necessary, in case other consumption con-
ditions can exist, that a prioritization mechanism is
in place. The address details are necessary to send
the invoice to the service consumer. Future market-
places have to consider these findings and implement
them accordingly to satisfy the clearing requirement
pointed out by the literature and the interviewed ex-
perts.
The GetTransactionData response is presented in
Figure 7 on the next page. The response contains de-
tails about a transaction’s status, operation and pro-
cessing time. The time is important because it indi-
cates to which time slot a transaction belongs and how
it has to be charged. A general usage price is charged
if a transaction does not fit into any special time slot.
A transaction can maximal fit into one special time
slot. All prices are taken from the contract to which
a transaction belongs. A transaction is free of charge
if it was not processed successfully. These character-
istics have been identified during the implementation
and particularly during the testing. Test data was de-
fined and the tests executed without any human in-
teraction. Tests were executed to check whether the
identified interfaces and their design is valid and ful-
fill the requirements of individual service clearing or
if certain parameters are still undetected. Furthermore
did the tests show what data has to be available and
accessible and what is the data access sequence.
5 CLEARING SERVICE
PROTOTYPE
Visual Rules, a modeling framework for Business
Rules Management (BRM) has been used to develop
a service for service clearing. The service runs as a
web service within a rule execution engine which it-
self runs on Apache Tomcat.
The clearing service uses and evaluates the capa-
bilities and the completeness of the introduced in-
terfaces for service clearing. It accesses automati-
CLOSER 2016 - 6th International Conference on Cloud Computing and Services Science
34
C:\Bosch-SI\Promotion\LaTeX\Dokumente\Konferenzen\7thIFIP_InternationalConferenceOnNewTechnologiesMobilityAndSecurity\ServiceCharging\pictures\GetContractDataResp.txt
Sonntag, 14. Juni 2015 09:31
<GetContractDataResp>
<ContractDetails>
<Contract>
<ContractId>20003</ContractId>
<ServiceName>ClearingService
</ServiceName>
<CreationDate>2015-01-15T11:36.12
</CreationDate>
</Contract>
<ConsumerAddress>
<Name>EMP</Name>
<Address>
<...>
</Address>
</ConsumerAddress>
<ProviderAddress>
<Name>CPO</Name>
<Address>
<...>
</Address>
</ProviderAddress>
<Pricing>
<UsageBasedPlan>
<IsDeleted>false</IsDeleted>
<BasicPrice>0</BasicPrice>
<UsagePrice>0.2</UsagePrice>
<TimePeriod></TimePeriod>
<Currency>EUR</Currency>
<TimeSpecification>
<startDay>Monday</startDay>
<endDay>Tuesday</endDay>
<startTime>09:00</startTime>
<endTime>20:00</endTime>
<usagePrice>0.1</usagePrice>
</TimeSpecification>
</UsageBasedPlan>
</Pricing>
</ContractDetails>
</GetContractDataResp>
-1-
Figure 6: Structure of GetContractData response.
cally internal marketplace data from outside as a third
party. The service clearing process complies to the
sequence presented in Figure 4. The service’s core
consists of an domain independent clearing algorithm.
Domain independent means that this algorithm does
not differentiate or evaluate the content of a provided
contract or transaction. As long as the provided data
passes the interface schema, the algorithm can calcu-
late the total costs for any kind of service transaction
that has been processed via the marketplace. With-
out the access at transaction and contract data, a third
party clearing operator would not be able for doing so
via a marketplace. The algorithm is able to process
any number of special time slots while considering
only transactions which have been processed success-
fully. The developed algorithm is presented below.
Let S be the set of all relevant service consump-
tions:
S =
x | x is service consumption
C:\Bosch-SI\Promotion\LaTeX\Dokumente\Konferenzen\7thIFIP_InternationalConferenceOnNewTechnologiesMobilityAndSecurity\ServiceCharging\pictures\GetTransactionDataListResp.txt
Donnerstag, 11. Juni 2015 14:50
<GetTransactionDataResp>
<TransactionDetails>
<Operation>STARTCHARGE</Operation>
<RequestStatusCode>
<sessionStatusName>Success</sessionStatusName>
</RequestStatusCode>
<RequestTime>2015-05-15T17:04:47</RequestTime>
<ResponseTime>2015-07-15T17:04:54</ResponseTime>
<RequestStatusCode>
<sessionStatusName>Success</sessionStatusName>
</RequestStatusCode>
</TransactionDetails>
<TransactionDetails>
<Operation>STARTCHARGE</Operation>
<RequestStatusCode>
<sessionStatusName>Success</sessionStatusName>
</RequestStatusCode>
<RequestTime>2015-05-18T13:38:32</RequestTime>
<ResponseTime>2015-05-18T13:38:39</ResponseTime>
<RequestStatusCode>
<sessionStatusName>Success</sessionStatusName>
</RequestStatusCode>
</TransactionDetails>
<TransactionDetails>
<Operation>STARTCHARGE</Operation>
<RequestStatusCode>
<sessionStatusName>Success</sessionStatusName>
</RequestStatusCode>
<RequestTime>2015-05-19T19:59:55</RequestTime>
<ResponseTime>2015-05-19T20:00:02</ResponseTime>
<RequestStatusCode>
<sessionStatusName>Success</sessionStatusName>
</RequestStatusCode>
</TransactionDetails>
</GetTransactionDataResp>
-1-
Figure 7: GetTransactionData response data with transac-
tion status, type and times.
For s S is
status(s) :=
successful if s successful
failure if s failure
That defines a function
status:
S
successful, failure
s 7− status(s)
For s S let t
s
be the timestamp (ts) that corresponds
to the service consumption s. That defines a function
ts:
S
t | t is ts
s 7− t
s
T is the set of special time slots specified in the con-
tract:
T =
Z | Z is special time slot
Now let P be a function that maps special time slots
to related prices.
P:
T R
0
=
r R | r 0
Z 7− P(Z)
The final formula TC is thus
TC = BP · χ (successful)
{
status(s) | sS
}
+
ZT
sS
suc
χ (
Z
ts(s))·
Individual Service Clearing as a Business Service: A Capable Reference Solution for B2B Mobility Marketplaces
35
P(Z) +
sS
suc, T
c
UP
whereby:
S
suc
=
s S : status(s) = successful
S
suc, T
c
=
s S
suc
| Z T : ts(s) 6∈ Z
All successful transactions are checked whether they
fit into a special time slot or not. The respective usage
costs are accumulated. If at least one transaction is
successful then the basic usage price is charged too.
The total costs is the sum of all individual transaction
costs plus the basic usage price. Figure 8 depicts the
output of the service-clearing service. The provided
data is the one presented in Figure 6 and Figure 7.
C:\Bosch-SI\Promotion\LaTeX\Dokumente\Konferenzen\7thIFIP_InternationalConferenceOnNewTechnologiesMobilityAndSecurity\ServiceCharging\pictures\transactionList.txtDonnerstag, 11. Juni 2015 18:43
<transactionList>
<entry>
Transaction: ChargeAuthorizationStart from:
Friday, 2015-05-15 17:04:47 costs: 0:20
</entry>
<entry>
Transaction: ChargeAuthorizationStart from:
Monday, 2015-05-18 13:38:32 costs: 0:10
</entry>
<entry>
Transaction: ChargeAuthorizationStart from:
Tuesday, 2015-05-19 19:59:55 costs: 0:10
</entry>
</transactionList>
<totalCost>0.40</totalCost>
<numberTransaction>3</numberTransaction>
<numberSucTransactions>3</numberSucTransactions>
-1-
Figure 8: Output of the developed service for service-
clearing.
The value of numberTransaction in Figure 8 in-
dicates how many transactions have been processed
by the algorithm. The number of transactions with
are actually charged, based on its status, are given by
numberSucTransactions. Both values can be either
equal or numberTransaction is higher than number-
SucTransactions. The transactionList in Figure 8 de-
picts all processed transactions. This overview is pro-
vided within the invoice. It shows that different prices
for the transactions have been charged. Two transac-
tions fit into a special time specification and one does
not. Therefore two transactions are charged with a
special price (0.10 Euro) and one with the general us-
age price (0.20 Euro). Because the basic usage price
is zero the total cost for the charged transactions is
0.40 Euro. Furthermore does the transactionList de-
pict, that an overall number of three transactions have
been processed and three transactions are considered
for the total costs calculation.
Different test sets have been executed to demon-
strate the algorithms performance. The test results are
shown in Table 2.
Table 2: Rule model and algorithm performance measure-
ment.
Processed
Data Sets
Time
Slots
Execution
Time
Read
Data
500 3 825-845 ms 222 ms
5 980-1000 ms
10 1045-1060 ms
500 3 1200-1270 ms 360 ms
5 1300-1400 ms
10 1400-1450 ms
2000 3 1600-1610 ms 572 ms
5 1710-1735 ms
10 1790-1825 ms
3000 3 1930-2020 ms 703 ms
5 2120-2190 ms
10 2180-2360 ms
Because the test performance depends on the un-
derlying IT infrastructure and internet connection, a
scenario has been chosen that assumes that a elec-
tronic marketplace hosts the clearing service. Trans-
action data is not queried from a database somewhere
in the network but from a local CSV file. This avoids
that the infrastructure has a deep impact on the tests.
The transaction data column shows how many test
transactions are processed by the rule engine and the
time slots shows how many special time slots have
been checked against the transactions. The execu-
tion time shows the complete execution time of the
rule model, inclusive reading the test data, validate
the data and compute the price algorithm. The Read
Data column indicates how long it the clearing pro-
cess took to read the test data.
6 CONCLUSION
The test cases demonstrated that the elaborated clear-
ing service processes correctly. This implies that
the interface design, implementation and execution
sequence are valid. Therefore, data access across
system boundaries has been achieved. The tests
have confirmed our identified data which has to be
accessed and combined properly to achieve service
clearing at all. For the individual service clearing,
the tests have shown that it is necessary to entitle cer-
tain roles to access and process third party data. This
functionality is key and without no individual and out-
sourced service clearing is possible via a marketplace.
Furthermore did the tests proof that role-based driven
interfaces are suitable for the problem context and
CLOSER 2016 - 6th International Conference on Cloud Computing and Services Science
36
provide a powerful mechanism to enable independent
interface consumption. The paper demonstrated how
individual service clearing can be achieved by out-
sourcing it to respective marketplace participants. As
pointed out by the experts, to the current day all avail-
able B2B service marketplaces in the electric mobility
domain lack individual and outsourced service clear-
ing. With the presented solution on hand, they can
start implementing this urgently required functional-
ity. They can use the presented reference implemen-
tation as a foundation to close this gap. Overcoming
service marketplaces’ system boundaries to access all
kind of stored data is a step towards marketplace in-
terconnectivity (Strasser and Albayrak, 2016).
This functionality is already available for flight or
hotel portals, but in the mobility domain the capabil-
ity for data exchange among different systems is still
fiction. One of the problems why it is not yet possible
to interconnect mobility systems is, according to the
interviewed experts, that (electric) mobility services
and marketplaces are sill under development and have
no mass market. The experts point out that companies
hesitate to invest money, do not open their systems
and do not tightly collaborate with each other. The
developed and presented interfaces have enabled the
required integration and provision of individual and
outsourced service clearing. They can also be used as
a foundation for future interface implementations.
7 OUTLOOK ON FUTURE WORK
As Buyya et al. (2008) recognized, protocol exten-
sion is key for future service interactions. The de-
signed interfaces can be used as a foundation to start
discussions about a standard for B2B service clear-
ing in service provisioning marketplaces. The expe-
rience gained during the research and the interface
implementation will contribute to discussions about
required interfaces for marketplace interconnectivity.
This research area is supposed to be a topic which
deserves high attention as it is of high urgency for
shifting today’s way of mobility. Once a marketplace
provides the suggested interfaces for service clear-
ing and interconnectivity between different market-
places is not fiction anymore, the proposed clearing
approach might be used as a foundation to enable the
clearing of service requests which have been roamed
between fragmented service marketplaces. For this
research we will build on the proposed interconnec-
tivity approach elaborated by (Strasser and Albayrak,
2016) as this already enables the exchange of certain
data sets among fragmented service marketplaces via
an appropriate and reliable connection management.
ACKNOWLEDGEMENT
This work has been conducted within a project (sup-
port code 16SBB007C ) funded by the German Fed-
eral Ministry of Economics and Technology.
Interview partners were from Bosch SI, IBM
R&D, T-Systems, Siemens, Smartlab GmbH, EnBW,
BridgingIT, Agency for Electric Mobility, Berlin,
Federal Highway Research Institute (BASt), Depart-
ment of Traffic and Infrastructure (BW), University
of Applied Science, Ludwigshafen (all Germany),
Gireve (France), Proeftuin-EV - Flemish Institute for
Technological Research (Belgium), Bosch SI (Singa-
pore/Asia). The interviews have been conducted be-
tween July - August 2015.
REFERENCES
˚
A gerfalk, P., Bannon, L., and Fitzgerald, B. (2006). Action
in Language, Organisations and Information Systems.
Number March.
Balocco, R., Perego, A., and Perotti, S. (2010). B2b eMar-
ketplaces Aclassification framework to analyse busi-
ness models and critical success factors. Industrial
Management & Data Systems, 110(8):1117—-1137.
Buchinger, U., Lindmark, S., and Braet, O. (2013). Busi-
ness Model Scenarios for an Open Service Platform
for Multi-Modal Electric Vehicle Sharing. In SMART
2013: The second International Conference on Smart
Systems, Devices and Technologies, pages 7–14.
Buyya, R., Yeo, C., and Venugopal, S. (2008). Market-
oriented cloud computing: Vision, hype, and reality
for delivering it services as computing utilities. In
Proceedings - 10th IEEE International Conference on
High Performance Computing and Communications,
HPCC 2008 (2008), pages 5–13. IEEE.
CROME (2014). CROME. http://crome-project.eu/. Last
accessed on 2014-05-15.
Ghenniwa, H., Huhns, M. N., and Shen, W. (2005). eMar-
ketplaces for enterprise and cross enterprise integra-
tion. Data & Knowledge Engineering, 52(1):33–59.
Green-eMotion (2015). Green eMotion.
http://www.greenemotion-project.eu/. Last accessed
on 2015-11-05.
Grieger, M. (2003). Electronic marketplaces: A literature
review and a call for supply chain management re-
search. European Journal of Operational Research,
144(2):280–294.
Gubbi, J., Buyya, R., Marusic, S., and Palaniswami, M.
(2013). Internet of Things (IoT): A vision, architec-
tural elements, and future directions. Future Genera-
tion Computer Systems, 29(7):1645–1660.
Heinrich, B., Huber, A., and Zimmermann, S. (2011).
Make-and-Sell or Buy of Web Services. In 19th Eu-
ropean Conference on Information Systems (ECIS).
Hubject (2015). Hubject. http://www.hubject.com/
?lang=en. Last accessed on 2015-11-5.
Individual Service Clearing as a Business Service: A Capable Reference Solution for B2B Mobility Marketplaces
37
Johnson, M. (2013). Critical success factors for B2B e-
markets: a strategic fit perspective. Marketing Intelli-
gence & Planning, 31(4):337–366.
Legner, C. (2007). Is there a Market for Web Services? -
An Analysis of Web Services Directories. In Service-
oriented computing-ICSOC 2007 workshops, pages
29—-42. Springer.
Luttge, K. (2001). E-Charging API: Outsource Charging
to a Payment Service Provider. Intelligent Network
Workshop, 2001 IEEE, 00(C):216–222.
Multicity (2015). Multicity. https://www.multicity-
carsharing.de/en/. Last accessed on 2015-11-05.
Olympus (2015). Olympus. http://www.proeftuin-
olympus.be/en/home-1.htm. Last accessed on 2015-
11-05.
Parku (2015). parku. https://parku.ch/?lang=en. Last ac-
cessed on 2015-11-05.
Pfeiffer, A. and Bach, M. (2014). An E-Clearinghouse for
Energy and Infrastructure Services in E-Mobility. In
Operations Research Proceedings 2012, pages 303–
308. Springer International Publishing.
Rust, R. T. and Zahorik, A. J. (1993). Customer Satisfac-
tion, Customer Retention, and Market Share. Journal
of retailing, 69(2):193–215.
smartlab Innovationsgesellschaft mbh. smartlab. http://
smartlab-gmbh.de/. Last accessed on 2015-05-23.
Strasser, M. and Albayrak, S. (2015). Conceptual Architec-
ture for Self-Discovering in Fragmented Service Sys-
tems. In 7th International Confrence on New Tech-
nologies, Mobility and Security (NTMS), 2015, Paris.
Strasser, M. and Albayrak, S. (2016). Smart City Refer-
ence Model : Interconnectivity for On-Demand User
to Service Authentication. In 12th International Con-
ference on Industrial Engineering (ICIE) (to be pub-
lished).
Strasser, M., Weiner, N., and Albayrak, S. (2016). Out-
sourced Invoice Service : Service-Clearing as SaaS
in Mobility Service Marketplaces. In Network Oper-
ations and Management Symposium (NOMS) (to be
published).
Streetlife (2015). Streetlife. http://www.streetlife-
project.eu/index.html. Last accessed on 2015-11-05.
Thitimajshima, W., Esichaikul, V., and Krairit, D. (2015).
Developing a Conceptual Framework to Evaluate Pub-
lic B2B E-Marketplaces. In Proceedings of the Pacific
Asia Conference on Information Systems 2015.
Turban, E., King, D., Lee, J., Liang, T.-P., and Turban, D. C.
(2015). Electronic Commerce - Business-to-Business
E-Commerce. Springer.
Vidal, N., Iaarakkers, J., Marques, R., Scuro, P., Caleno,
F., Matrose, C., and Bolczek, M. (2011). Report on
Billing and stakeholders architecture and ICTs recom-
mendations. Technical Report 241295, G4V.
CLOSER 2016 - 6th International Conference on Cloud Computing and Services Science
38