Exploring Blockchain Technology to Improve Multi-party Relationship

in Business Process Management Systems

Paulo Henrique Alves

, Ronnie Paskin

, Isabella Frajhof

, Yang Ricardo Miranda

ao Gabriel Jardim

, Jose Jorge Brum Cardoso

, Eduardo Henrique Haddad Tress

Rog

erio Ferreira da Cunha

, Rafael Nasser

and Gustavo Robichez

Software Engineering Laboratory, Department of Informatics, Pontiﬁcial University Catholic of Rio de Janeiro, RJ, Brazil

Petrobras, Rio de Janeiro, Brazil

{jjbrum, eduardotress, roger}@petrobras.com.br

Keywords:

Business Process Management System, Workﬂow, Blockchain, Smart Contract, Architecture.

Abstract:

Business Process Management Systems (BMPSs) are often used to track activities, identify inefﬁciencies, and

streamline the workﬂow. Typically, BPMSs are used by a single organization for internal users and processes

through a trusted central system. However, scenarios involving multiple parties present a new challenge: to

ensure the reliability of registered information and strict adherence to business rules for all participants without

a central authority. Therefore, we explored the use of blockchain technology in association with a BPMS to

create a Distributed Business Process System (dBPMS). This integration can fulﬁll the requirements above

mentioned, creating tamper-proof registries, allowing reliable self-execution through smart contracts, within

a trusted environment for all parties, without the need for inter-party trust. The proposed solution provides a

workﬂow encompassing all activities and parties in an efﬁcient ecosystem.

1 INTRODUCTION

As a basic process modelling approach, workﬂows

usually focus on the sequence of events. Thus, Busi-

ness Process Management Systems (BPMSs) (Kara-

giannis, 1995) play a critical role to describe and track

activities in the organizational environment. How-

ever, it loses the connection between the model and

the developed platform as well as ignores the informa-

tional perspective (Hull et al., 2016). In order to im-

prove the big picture of organizational business pro-

cesses and the information traceability during the pro-

cess execution, we have integrated to the BPMS the

blockchain technology (BCT) in order to: (i) keep the

historical records; (ii) guarantee that data is tamper-

proof, and (iii) ensure that the processes are running

precisely as planned according to the deal made be-

tween the involved parties.

In this sense, mapping the business process is cru-

cial to guarantee the orchestration between business

goals and their execution. Therefore, it is vital to

check, and even redesign, the processes regularly, due

to the constant need to adapt businesses workﬂows

to attend to new demands. Furthermore, this orches-

tration often involves external entities (other compa-

nies), such as supply providers and regulatory entities.

Therefore, to ensure the reliability of registered infor-

mation and strict adherence to business rules for all

participants it is crucial to use proper technology.

Blockchain technology presents many features to

support such BPMS requirements, such as resources

to deal with untrusted parties, tamper-prooﬁng and

traceability of information. Furthermore, the use

of blockchain technology to distribute BPMS is a

novel approach never presented before. Moreover,

Blockchain Smart Contracts (BSCs) play a crucial

role in this environment, enabling self-executed con-

tracts that use programming language code, which do

not depend on parties will to be enforced. Once de-

ployed, the smart contract code cannot be undone, an

effect derived from the immutability of BSC.

Usually, blockchain enterprise solutions are ex-

ecuted in a permissioned (i.e. private) blockchain,

where the information is shared with a particular

group of stakeholders, which is a speciﬁc feature of

this kind of blockchain. Unlike permissionless (i.e.

public) blockchains, the permissioned blockchains al-

low faster and cheaper transactions because of their

limited number of participants, allowing parties to

Alves, P., Paskin, R., Frajhof, I., Miranda, Y., Jardim, J., Cardoso, J., Tress, E., Ferreira da Cunha, R., Nasser, R. and Robichez, G.

Exploring Blockchain Technology to Improve Multi-party Relationship in Business Process Management Systems.

DOI: 10.5220/0009565108170825

In Proceedings of the 22nd International Conference on Enterprise Information Systems (ICEIS 2020) - Volume 2, pages 817-825

ISBN: 978-989-758-423-7

817

rapidly achieve consensus in order to approve trans-

actions.

In this context, this paper proposes an exploratory

study to analyze and create a new solution using

BPMS and blockchain technology in an integrated

manner. To implement such solution, twelve different

workﬂow tools were identiﬁed and analyzed, whereas

only one was selected to develop a blockchain inte-

grated solution enforced by smart contracts. To ana-

lyze and demonstrate the beneﬁts of this solution, we

have exempliﬁed the usage of this architecture, apply-

ing it to a real processes at an Oil and Gas company.

The paper is structured as follows. Section 2 fo-

cuses on the background, while Section 3 discusses

related work. Section 4 presents the comparison of

BPMS existing projects and Section 5 presents the

proposed solution for distributed BPMS (dBPMS),

and Section 6 shows an application scenario. Finally,

Section 7 presents the conclusions and future perspec-

tives.

2 BACKGROUND

This section describes the main concepts used in this

paper, such as Business Process Management suits

and blockchain technology. Both concepts are fun-

damental to understand how they are used together to

create a dBPMS, and the beneﬁts of this solution.

2.1 Business Process Management

Systems

In general, Business Process Management orchestrate

the organizational resources, informational structures

and business goals (Hull et al., 2016). Moreover,

BPM regulate inter-organizational relationships, and

guarantee the execution of tasks and decisions that

yield an outcome that adds value for an enterprise

(Dumas et al., 2013).

Workﬂows are typically applied for process

modelling approach and business coordination (van

Der Aalst, 2009). They usually are focused on con-

trol ﬂows; however, in these cases, the informational

perspective is ignored. As a result, instead of improv-

ing operational innovation, business agents frequently

focus on what should be done as a process, instead

of aiming at issues such as the data security, trans-

parency, and data sharing. (Bhattacharya et al., 2007).

Based on this perspective, many suites were devel-

oped to deliver resources to improve business process

management, such as presented in (Abdelgader et al.,

2013; Van Der Aalst and Ter Hofstede, 2005; Cha-

banoles and Ozil, 2015; Rademakers, 2012; Cumber-

lidge, 2007; Fernandez, 2013).

Business Process Management Systems are soft-

ware tools used to allow managing business process

life cycle (Meidan et al., 2017). Moreover, they are

also used as a uniﬁed solution to model, execute and

monitor business processes. However, each solution

present different features, and deciding which one is

appropriate to a particular situation can be challeng-

ing. Besides, BPMS often proposes a centralized so-

lution and no features to deal with the lack of trust

and transparency between different parties in a spe-

ciﬁc environment.

Therefore, in the following sections, we propose a

new BPMS model to distribute the information with-

out a central authority, aiming not only to improve

the execution and monitoring processes, but also to

store tamper-proof information. This new approach

also improves the process’ data and transparency to

enhance trust for involved parties.

2.2 Blockchain Technology

The blockchain technology was introduced in the Bit-

coin white paper (Nakamoto, 2009) as a technol-

ogy to ensure transparency of the registered data,

dismisses inter-party trust, and guarantee data im-

mutability. Consensus mechanism play an essential

role on the validation of the registered transactions,

which inﬂuences on energy consumption and network

performance. Later, other kinds of distributed solu-

tions used this technology to enjoy all of blockchain

features, such as BSC, for instance.

Furthermore, there are two different types of

blockchain: (i) permissionless, and (ii) permis-

sioned blockchains (Peters and Panayi, 2016) (Alves

et al., 2018). The former is the original purposed

blockchain, with no imposed limitation of the agents

that can access the network or its content, maintain-

ing public all registered data, which is distributed

to the networks’ node. Conversely, permissionless

blockchains only allow information to be shared with

selected entities, in which access to data can either

be public or private to the participants, according to

previous combination. The permissioned blockchains

usually presents better performance than the public

ones due to the smaller number of participants in-

volved, requiring less effort on achieving a consensus.

Based on the blockchain technology, smart con-

tracts are a representation of a contract, roles, agree-

ments and obligations written on programming lan-

guage, which are coded throughout simple if-then-

else logic (Szabo, 1994) in a distributed network.

It enables auto-execution of the negotiated obliga-

ICEIS 2020 - 22nd International Conference on Enterprise Information Systems

818

tions, without the need of a central authority to en-

force the agreed terms; oracles present an impor-

tant role in this case (Adler et al., 2018). Thus, un-

trusted relationships can rely on the blockchain en-

vironment to guarantee transparency between partici-

pants, self-execution of the agreed terms and reliabil-

ity of the registered data, due to its consensus mecha-

nism. Also, trust is reinforced since data is shared be-

tween the network participants and cannot be altered

by them. Besides, smart contract platforms allows

the development of distributed applications (dApp),

which are a kind of internet application where the

backend runs on distributed, or decentralized, peer-

to-peer network. Also, no single participant has com-

plete control over the dApp on the network (Prusty,

2017).

Therefore, in the following sections we explain

our Distributed Business Process Management Sys-

tem (dBPMS) proposal.

3 RELATED WORK

This section presents some related works regarding

BPM and blockchain technology. They were cate-

gorized as follow: (i) blockchain challenges and op-

portunities in the BPM context; (ii) Business Process

Modeling Notation Systems integration with BCT

to redesign governance; (iii) BPM traceability using

blockchain, (iv) BPMS tools, and (v) BPMN usage

for BSC modelling and execution.

The authors in (Mendling et al., 2018) present

some challenges and opportunities for blockchain ab-

stractions and heuristics development, presenting a

promising direction from BPMN choreography dia-

grams and data-aware workﬂows. The combination

of these two perspectives, data and process, seems

to be interesting for the top-down design of multi-

organizational processes. Moreover, as blockchain

technology, among others, delivers immutability, data

sharing management (permissioned blockchains) and

no central authority, the authors cited the impor-

tance of process redesign to consider all these fea-

tures while developing a blockchain-based solution.

However, (Mendling et al., 2018) only highlights the

discussion, and does not present any proposals.

Moreover, the authors in (Viriyasitavat et al.,

2018) argues that the main bottleneck in BPM sys-

tems is the trust management, lack of transparency

and evaluation. They also present three main beneﬁts

of the use blockchain technology application in BPM:

(i) trust establishment by consensus mechanism and

transparency; (ii) costs reduction due to the absence

of a central authority, and (iii) transaction automation

by real-time update and veriﬁcation of the quality of

service. They developed a BPM blockchain-based ap-

proach where a Solidity smart contract represents the

business processes. Although the authors created a

framework from this approach, it does not guarantee

that the process will be executed as planned in the

modelling phase.

Some approaches (Garc

ıa-Ba

nuelos et al., 2017;

opez-Pintado et al., 2019; Weber et al., 2016; Di Ci-

ccio et al., 2019), have proposed frameworks and

tools to translate BPMN in a BSC. For instance,

Caterpillar (L

opez-Pintado et al., 2019) and Lorikeet

(Di Ciccio et al., 2019) are tools that are used to

model business process and to automatically gener-

ate an Ethereum smart contract structure. Further-

more, the authors in (Garc

ıa-Ba

nuelos et al., 2017)

and (Di Ciccio et al., 2019) apply the same approach

using Petri net (Murata, 1989) to translate the gen-

erated BPMN process model. Although those works

have developed an Ethereum smart contract in com-

pliance with the BPMN process model, the authors

do not present a relation between the modelled pro-

cess and the smart contract after its ﬁrst execution. In

other words, once created, the smart contract is not

affected by the model anymore. Unlike their solu-

tion, we developed a persistence mode using a private

blockchain; hence, the modelled process will always

generate data persistence on the blockchain.

Furthermore, the authors in (Meidan et al., 2017)

and (Garc

es et al., 2009) presents the result of a con-

ducted survey which organizations evaluated the ben-

eﬁts of open source BPMS systems. The former pa-

per presents a detailed description of each platform,

describing their modelling, design, deployment, exe-

cution and monitoring. The latter presents a compar-

ative framework based on the Workﬂow Management

Coalition. Although the authors proposes a general

feature evaluation, many essential aspects were not

considered by them, such as process interoperability,

document management and process validation, for in-

stance. Furthermore, both surveys indicates the im-

portance of quality and optimization improvement in

the BPMS environment.

Finally, Prybila et al. (Prybila et al., 2017) and

Meroni et al. (Meroni et al., 2019) present the

blockchain technology as a provider of BPMS decen-

tralization and BPM data distribution, respectively.

The former presents choreographies when the con-

trol over process instance is shared between different

companies or organizations. Moreover, the authors

developed a runtime veriﬁcation for choreographies

by the usage of the Bitcoin blockchain. However,

the Bitcoin blockchain presents limited features in

comparison with smart contract blockchain platforms;

Exploring Blockchain Technology to Improve Multi-party Relationship in Business Process Management Systems

819

thus, the authors lost the compliance checker and the

self-executed features derived from the blockchain

smart contract. In its turn, the latter presents an

Ethereum-based solution to store documents and IoT

data on the blockchain to improve trust in monitoring

artifacts (physical objects) in a multi-party business

process. However, the authors did not approach the

importance of assuring business data privacy policy

in a permissionless blockchain.

Thus, many different approaches are proposing

solutions related to the usage of blockchain and BPM

to improve transparency, data availability, and to help

with the mapping of the process and its functions on

smart contracts. However, a solution presenting a dis-

tributed business process management system is still

missing. In this sense, we identiﬁed this opportunity

and developed an approach where all the information

is legitimately stored. Moreover, we present an appli-

cation example of our solution in order to demonstrate

precisely its execution in the Oil and Gas scenario.

4 BPM WORKFLOW

PLATFORMS COMPARISON

This section describes the evaluated BPMSs and why

we selected one of the investigated workﬂow tools

to develop the solution to integrate it with a private

blockchain. For this task, we analyzed twelve dif-

ferent platforms, among open-source and commercial

solutions, looking for the following requirements: (i)

business process modeling tool; (ii) workﬂow execu-

tion with forms support; (iii) API availability; (iv)

open-source project and (v) programming language

for open-source projects,

First, the modeling tool must understand and or-

chestrate business processes; this is the initial step to-

wards the construction of a dBPMS. Then, in order to

execute the modeled processes, a workﬂow execution

in compliance with the modeled processes is needed;

besides, this workﬂow needs to support manners to

add data directly on the platform during its processes

execution. Next, the API is required to send infor-

mation to a private blockchain. In this manner, we

selected a set of platforms presented on our related

work and analyzed them in order to make a well-

based decision regarding which platform we should

use to develop our architecture. Moreover, open-

source projects have increasingly presented a grow-

ing impact on the software industry and have often

become primary competitors to commercial software

(Fitzgerald, 2006; Bonaccorsi and Rossi, 2003).

Table 1 presents a comparison between all the

analysed platforms, i.e. commercial licenses and

open-source. All platforms presented an API REST

to integrate services and systems to their solutions.

Moreover, ﬁve of seven open-source projects pre-

sented all the ﬁve required features (Farah et al.,

2014). As a commercial solution, Pipefy, Wrike,

Monday.com, Kissﬂow, and Integrity do not present

details of their solution and the language code. Thus,

we decided to remove these projects from our set of

possibilities.

Table 1: BPM Systems Evaluation.

Tool/ Feature BPMN Forms OpenSource

Pipefy NO NO NO

Wrike NO YES NO

Monday.com NO YES NO

Kissﬂow YES YES NO

Integrity YES NO NO

Joget NO YES YES

YAWL NO YES YES

Bonita YES YES YES

ProcessMaker YES YES YES

Activiti YES YES YES

jBPM YES YES YES

Camunda YES YES YES

We present below an overview of the ﬁve projects

that fulﬁlled our ﬁve requirements above-mentioned,

describing more details of each open-source project.

Bonita delivers BPM automation and application

projects, and is composed of two parts: (i) BPM Stu-

dio and (ii) BPM Platform. The BPM Studio is a

graphical interface for process and forms develop-

ment; whereas the BPM Platform handles business

process execution and Bonita Portal. The latter allows

visualizing tasks and tracking user activity.

ProcessMaker is based on three main compo-

nents: (i) business process design; (ii) workﬂow en-

gine, and (iii) adapters. The BP design presents tools

to map processes and business rules. Moreover, it

also allows the creation of forms and the attachment

of documents as input to the process. The engine is

used to automate processes. Finally, the adapters are

a suite of connectors to integrating external solutions.

Activiti, as all the open-sources projects pre-

sented, Activity allows users to design BPMN 2.0

environments. The process engine was developed in

Java, and distributed under the Apache license. It is

composed of a modeler, and Activiti engine to execute

the modeled processes.

jBPM is also an open-source project written in

Java. It allows the design, execution and monitoring

of the business process life cycle. Moreover, it also al-

lows the user to model business process throughout a

web-based interface. Remote APIs and tasks services

are also available.

ICEIS 2020 - 22nd International Conference on Enterprise Information Systems

820

Camunda delivers process automation by a BPM

java-based framework, and consists of two parts: (i)

Camunda Modeler and (ii) Camunda Engine. The for-

mer is a web-based application to model business pro-

cesses. The latter is an engine composed of the fol-

lowing components: Cockpit, Tasklist, Workﬂow En-

gine, and Decision Engine, which are used to execute

and monitor processes.

Although ﬁve out of seven open-source projects

attended to our requirements, we had to choose only

one of them, even though we acknowledge that any

of the other four projects could also be used. In this

sense, we have chosen Camunda, because this project

presents an architecture that supports process appli-

cation event listeners, classiﬁed as (i) task event lis-

tener and (ii) execution event listeners. The former

allows reacting to task events, such as creation, as-

signment and conclusion, while the latter allows re-

acting to trigger events at the same time that a move

is executed further on the diagram.

Furthermore, the Camunda BPMN model API en-

ables the extraction of data from existing process def-

inition. Likewise, Camunda Process Engine guar-

antees that Events are delegated to the right process

application. The combination of these features cre-

ated an adequate environment to integrate Camunda

BMPS with blockchain technology, allowing for the

registration of immutable data, high data availability

and permissioned access between organizations. In

the next sections, we explain in details the usage of

Camunda and its architecture when integrated with a

private blockchain, thus creating a dBPMS.

5 PROPOSED SOLUTION

This section describes the main concepts used to de-

velop the dBPMS architecture, and how such archi-

tecture will work in the business process manage-

ment environment with the integration of blockchain

technology, thus, guaranteeing transparency, tamper-

proof information and self-executed smart contracts.

This kind of architecture, used to store rules, and cur-

rent state of tasks in the blockchain trough BSC, was

also proposed in (Smirnov et al., 2019).

5.1 Camunda BPM

Camunda BPM is a Java-based open-source platform

for Business Process Management automation. Its

stack is composed of the following applications:

• BPMN Workﬂow Engine: Used to execute

(micro-)service orchestration and human task

management based on the BPMN model;

• Modeler: Used to create BPMN process diagrams

and DMN decision tables and deploy these work-

ﬂows to the Camunda Engines;

• Tasklist: Used to perform human task manage-

ment that allows users to work on assigned tasks,

such as form ﬁlling;

• Cockpit: Used to monitor workﬂows and deci-

sions in production to identify, analyse and solve

technical problems.

Moreover, Camunda BPM also offers a REST API

to interact with the engines and allows the integration

with external applications. Besides, in order to han-

dle the workﬂow execution, Camunda BPM offers the

possibilities to write Delegation Code, a Java Code to

execute external applications, scripts or evaluate ex-

pressions when certain events occur during process

execution by the Execution Listener. It can attach

any event in the normal token ﬂow, for instance, the

start and end of a process instance, activity, gateway

or event.

The Execution Listener is an exter-

nal Java-class that should implement the

org.camunda.bpm.engine.delegate.ExecutionListener

interface. When the process, event, gateway or

activity occurs, the notify method notiﬁes the lis-

tener implementation to store the information in

the blockchain. It allows variables manipulation

and log creations of all completed steps by process

instances; thus, the listener implementation can call

a blockchain Hyperledger Fabric node and persists

JSON objects with all the variables of any process.

Therefore, the listener can make requests to external

systems and manipulate the variables within the

BPMN elements.

5.2 Hyperledger Integration

The integration of the proposed BPMN solution with

Hyperledger Fabric is rather simple, but still powerful

enough to bring forward the beneﬁts of a blockchain

distributed system.

The Hyperledger Fabric environment consists of

permissioned distributed nodes that execute smart

contracts. These contracts are speciﬁcally built to reg-

ister and manage information created by the BPMN

solution. The BPMN system communicates with the

blockchain nodes using Javascript clients and the ap-

propriate Hyperledger SDKs. These clients act as in-

termediaries to the blockchain network to allow inter-

actions with the smart contracts. All execution and

data generated from these interactions are distributed

to the blockchain network; hence, the stored data be-

comes immutable and auditable.

Exploring Blockchain Technology to Improve Multi-party Relationship in Business Process Management Systems

821

Moreover, Hyperledger Fabric is a permissioned

blockchain where smart contracts are executed on

controlled and access-restricted communication chan-

nels. In other words, only speciﬁc and authorized par-

ticipants can interact and access transactions on a par-

ticular channel. Thus, these channels can be used to

effectively maintain private information between par-

ticipants of the system, allowing for data governance

and provenance.

5.3 dBPMS Proposal

Business Process Management Systems (BPMSs) are

sophisticated tools that are able to model, execute, co-

ordinate and monitor business processes. BPMS aims

to handle the relationship between one or more par-

ties as a centralized system. However, BPMSs do

not provide any external guarantee of data integrity

or immutable log record. All BPMS listed (see sec-

tion 4, Table 1) used a traditional centralized database

to store the transmitted information. Therefore, we

propose a blockchain-based solution applying critical

blockchain features.

The integration between Camunda BPMS and Hy-

perledger private blockchain result in a trusted and

distributed BPMS. Figure 1 depicts the proposed so-

lution. Initially, the Camunda Modeler is used to de-

sign the BPMN 2.0 business model in a web-based

interface, available at any web browser. Then, the

Camunda BPM platform orchestrate and executes the

designed processes.

Furthermore, the BPM platform is composed of

two main steps. The ﬁrst step aims to generate tasks

and assign them to users. The second aims to exe-

cute tasks according to the designed business process

rules. The transmitted information is stored in a tra-

ditional database. Camunda supports MySQL, Mari-

aDB, Oracle, IBM DB2, PostgreSQL, Microsoft SQL

Server and H2 databases.

Camunda BPM developed a delegation code that

allows for the execution of external Java code, scripts

or the evaluation of expressions during the process ex-

ecution. There are four different types of delegation

code: (i) Execution Listeners, that can be attached to

any event within the normal token ﬂow; (ii) Java Dele-

gates, which can be attached to a BPMN Service Task;

(iii) Delegate Variable Mapping, that can be attached

to a Call Activity, and (iv) Task Listeners which can

be attached to events within the user task life cycle.

Thus, we applied the Execution Listeners to allow

the blockchain data persistence. Camunda Listener

is responsible for storing sensitive business informa-

tion on the Hyperledger blockchain. This ensures the

trustworthiness of registered information and strict

Figure 1: Distributed Business Process System.

adherence to business rules for all participants. Since

Hyperledger creates and executes blockchain smart

contracts in the distributed environment, there is no

need of a central authority to validate the registered

transactions on the blockchain.

Figure 2 depicts our Hyperledger architecture.

There are three different nodes types on our Hyper-

ledger Fabric network. The Orderer nodes maintain

consensus of all information in the system and they

are the pillar to the entire blockchain system. They

are responsible for organizing data creation and shar-

ing throughout the network. On the other hand, Peer

nodes maintain and execute smart contracts. They are

oriented by the Orderes to maintain consensus and

consistency of contracts data.

Figure 2: Hyperledger Architecture.

Our application uses smart contracts, that run on

these Peers, to register BPMS log data. Moreover,

Client nodes act as interfaces between the blockchain

ICEIS 2020 - 22nd International Conference on Enterprise Information Systems

822

Figure 3: Simpliﬁed Geophysics Service Solicitation Process.

and external systems. They are used in our solu-

tion to provide the link between the BPMS and the

blockchain. Clients interact directly with nodes to

consult, call and create smart contracts. All other

nodes, i.e., Orderers and Peers, can communicate

freely between themselves.

Furthermore, Hyperledger Fabric implements the

concept of communication channels. These channels

restrict smart contracts execution and data access to

speciﬁc Peer nodes. In our application, this is used

to maintain private information between participants

that need to interact, but do not want to expose all data

transacted to everyone in the network.

Hyperledger Fabric also allows simple integration

with various modern programming languages. The

used integration consists of an interaction between a

blockchain Peer node (through a Client node) that val-

idates transactions and creates smart contract calls,

and external applications. In this manner, the Ca-

munda Listener is able to execute smart contracts

commands in the Hyperledger Fabric network, per-

sisting, consulting and changing data. The blockchain

smart contract ensure self-execution of the agreed

obligation embedded on code. Moreover, this kind

of platform, private blockchains, also allows data

sharing, privacy management, and fast transactions,

which is often vital to the business. A positive aspect

of this application is that polices related to which in-

formation should, or not, be stored on the blockchain

can be deﬁned by each application context, following

the company’s security guidelines.

Moreover, in order to integrate other BPMSs, we

also have saved the information metadata generated

from the BPMS process execution in the blockchain,

the class HyperledgerLoggerListener.kt shows how it

was made

. This metadata allows further different

All the source code changed and created from scratch

BPMSs integration, since they will be able to query

the blockchain data and recreate the objects used in a

Camunda BPM Platform.

In summary, the decentralized backend and the

data immutability are important elements to assure

high availability, secure and transparent systems.

Combined with blockchain smart contracts, such sys-

tems are able to provide the ecosystem with these

characteristics.

6 USE CASE: GEOPHYSICS DATA

ACQUISITION PROCESS

The proposed example scenario was developed on the

Oil and Gas context, speciﬁcally the geophysics data

acquisition, which is a familiar scenario in this sector.

In order to evaluate the pros and cons to explore a

speciﬁc block, the geophysics department is requested

to evaluate the project availability.

The simpliﬁed scenario involves three actors: (i)

client, who is asking for the geophysics data; (ii) geo-

physics department, and (iii) the client purchasing de-

partment. The client and its purchasing department

are untrusted external entities, so all parties actions

need to be transparent, and all exchanged information

must be immutable and available for auditing.

To attend to these demands, the Geophysics De-

partment developed a standard workﬂow to deal with

external data requisition that guarantee data reliability

of registered information. Figure 3 depicts the sim-

pliﬁed business process created in Camunda Modeler.

The Camunda Tasklist was used to execute the mod-

elled process and allow users to perform their actions,

in this project are available in our repository. Repository

Link: shorturl.at/etyC1 (visited on 18/03/2020).

Exploring Blockchain Technology to Improve Multi-party Relationship in Business Process Management Systems

823

such as ﬁll up the forms and contracts.

As presented in Figure 3, the process starts when

a client request geophysics data, by ﬁlling up a form

with a detailed description of the wanted data. Within

seven days, the Geophysics Department must evaluate

such request and present in a workshop their solutions

to acquire the data. After the presentation, the client

has to ﬁll up the acceptance form with the selected so-

lution and conﬁrm their interest in pursuing the acqui-

sition data. Then, the Geophysics Department creates

a service contract and send it to the client purchasing

department. If all clauses comply with the requested

service, a ballot is started to get the acceptance of the

managers. Next, the ballot evaluation process is exe-

cuted by a simple voting smart contract system in two

weeks. If the managers decide to accept the contract,

the process will end. Otherwise, they must send the

rejection notes to the Geophysics Department to gen-

erate a new contract.

All the process information, i.e., forms, contracts,

and acceptance letters, were stored on the blockchain

in an encrypted manner, using many Hyperledger

Fabric nodes, some inside the organization, and oth-

ers outside, in the cloud. This procedure ensures data

transparency and auditability. Moreover, in the Cock-

pit, managers are able to monitor and follow the pro-

cess anytime.

Therefore, the developed scenario presents the

use of the proposed solution from modelling a pro-

cess to executing and storing the process data on the

blockchain. Any department inside the organization

can use the dBPMS solution, and more parties can be

added in this process, including external ones, with no

need to implement changes to the proposed architec-

ture. Furthermore, all data transmitted and stored are

immutable and available for the selected participants.

This is possible due to the permissioned blockchain

architecture.

7 CONCLUSION AND FUTURE

WORK

This paper proposes an approach to distributed BPMS

data through the use of blockchain technology to

improve multiple parties process orchestration, pro-

cess transparency, and allow tamper-proof registries.

Moreover, the proposed architecture also ensures the

reliability of business process information. In order

to select one BPMS to develop our solution, we an-

alyzed the BPMS state of art and selected the most

appropriate solution, according to our goals.

The main contributions of this research are: (i)

the BPM Systems evaluation, where we listed com-

mercial and open-source platforms, and (ii) the pro-

posal of a Distributed Business Process Management

System (dBPMS) to deliver blockhain features to the

business process management environment. The for-

mer is described in detail and an analysis of the core

business of each application is presented. The latter is

also described and exempliﬁed by a case study in the

context of the Oil and Gas industry, speciﬁcally, the

geophysics service data acquisition process.

As future work, we aim to develop a middle in-

terface layer between the Listener and the private

blockchain to allow different blockchain-base solu-

tion, instead of a speciﬁc one. Although the pre-

sented work allows the storage of information on the

blockchain, an interesting improvement would be the

other way around, where the BPMS request informa-

tion from the blockchain to rebuild the previously ex-

ecuted processes..

Another future work is related to the development

of a data visualization tool to apply process min-

ing techniques to improve the process performance.

Moreover, despite the model design, new modelling

entities considering the blockchain features could be

developed to improve the BPMN representation.

REFERENCES

Abdelgader, F. M. Z., Dawood, O. O. S., and Mustafa, M.

M. E. (2013). Comparison of the workﬂow manage-

ment systems bizagi, processmaker, and joget. In The

International Arab Conference on Information Tech-

nology (ACIT), Sudan University of Science and Tech-

nology, Khartoum, Sudan.

Adler, J., Berryhill, R., Veneris, A., Poulos, Z., Veira,

N., and Kastania, A. (2018). Astraea: A decen-

tralized blockchain oracle. In 2018 IEEE Interna-

tional Conference on Internet of Things (iThings) and

IEEE Green Computing and Communications (Green-

Com) and IEEE Cyber, Physical and Social Comput-

ing (CPSCom) and IEEE Smart Data (SmartData),

pages 1145–1152. IEEE.

Alves, P., Laigner, R., Nasser, R., Robichez, G., Lopes,

H., and Kalinowski, M. (2018). Desmistiﬁcando

blockchain: Conceitos e aplicac¸

oes. Computac¸

ao e

Sociedade. Rio de Janeiro: Sociedade Brasileira de

Computac¸

ao, pages 1–24.

Bhattacharya, K., Gerede, C., Hull, R., Liu, R., and Su, J.

(2007). Towards formal analysis of artifact-centric

business process models. In International Confer-

ence on Business Process Management, pages 288–

304. Springer.

Bonaccorsi, A. and Rossi, C. (2003). Why open source soft-

ware can succeed. Research policy, 32(7):1243–1258.

Chabanoles, N. and Ozil, P. (2015). Bonita bpm: An open-

source bpm-based application development platform

ICEIS 2020 - 22nd International Conference on Enterprise Information Systems

824

to build adaptable business applications. In Interna-

tional Conference on Business Process Management.

Cumberlidge, M. (2007). Business process management

with JBoss jBPM. Packt Publishing Ltd.

Di Ciccio, C., Cecconi, A., Dumas, M., Garcia-Ba

nuelos,

L., L

opez-Pintado, O., Lu, Q., Mendling, J., Pono-

marev, A., Tran, A. B., and Weber, I. (2019).

Blockchain support for collaborative business pro-

cesses. Informatik Spektrum, pages 1–9.

Dumas, M., La Rosa, M., Mendling, J., Reijers, H. A., et al.

(2013). Fundamentals of business process manage-

ment, volume 1. Springer.

Farah, G., Tejada, J. S., and Correal, D. (2014). Openhub: a

scalable architecture for the analysis of software qual-

ity attributes. In Proceedings of the 11th Working Con-

ference on Mining Software Repositories, pages 420–

423. ACM.

Fernandez, A. (2013). Camunda bpm platform loan assess-

ment process lab. Brisbane, Australia: Queensland

University of Technology.

Fitzgerald, B. (2006). The transformation of open source

software. MIS quarterly, pages 587–598.

Garc

es, R., de Jesus, T., Cardoso, J., and Valente, P. (2009).

Open source workﬂow management systems: a con-

cise survey. GWDL: A graphical workﬂow deﬁnition

language for business workﬂows. Springer Berlin Hei-

delberg.

Garc

ıa-Ba

nuelos, L., Ponomarev, A., Dumas, M., and We-

ber, I. (2017). Optimized execution of business pro-

cesses on blockchain. In International Conference

on Business Process Management, pages 130–146.

Springer.

Hull, R., Batra, V. S., Chen, Y.-M., Deutsch, A., Heath III,

F. F. T., and Vianu, V. (2016). Towards a Shared

Ledger Business Collaboration Language Based on

Data-Aware Processes. In Sheng, Q. Z., Stroulia, E.,

Tata, S., and Bhiri, S., editors, Service-Oriented Com-

puting, volume 9936, pages 18–36. Springer Interna-

tional Publishing, Cham.

Karagiannis, D. (1995). Bpms: business process manage-

ment systems. ACM SIGOIS Bulletin, 16(1):10–13.

opez-Pintado, O., Garc

ıa-Ba

nuelos, L., Dumas, M., We-

ber, I., and Ponomarev, A. (2019). Caterpillar: A

business process execution engine on the ethereum

blockchain. Software: Practice and Experience,

49(7):1162–1193.

Meidan, A., Garc

ıa-Garc

ıa, J. A., Escalona, M., and Ramos,

I. (2017). A survey on business processes man-

agement suites. Computer Standards & Interfaces,

51:71–86.

Mendling, J., Dustdar, S., Gal, A., Garc

ıa-Ba

nuelos, L.,

Governatori, G., Hull, R., Rosa, M. L., Leopold, H.,

Leymann, F., Recker, J., Reichert, M., Weber, I., Rei-

jers, H. A., Rinderle-Ma, S., Solti, A., Rosemann,

M., Schulte, S., Singh, M. P., Slaats, T., Staples, M.,

Weber, B., Weidlich, M., Aalst, W. V. D., Weske,

M., Xu, X., Zhu, L., Brocke, J. V., Cabanillas, C.,

Daniel, F., Debois, S., Ciccio, C. D., and Dumas, M.

(2018). Blockchains for Business Process Manage-

ment - Challenges and Opportunities. ACM Trans-

actions on Management Information Systems, 9(1):1–

16.

Meroni, G., Plebani, P., and Vona, F. (2019). Trusted

artifact-driven process monitoring of multi-party busi-

ness processes with blockchain. In International Con-

ference on Business Process Management, pages 55–

70. Springer.

Murata, T. (1989). Petri nets: Properties, analysis and ap-

plications. Proceedings of the IEEE, 77(4):541–580.

Nakamoto, S. (2009). Bitcoin: A peer-to-peer electronic

cash system. Technical report.

Peters, G. W. and Panayi, E. (2016). Understanding modern

banking ledgers through blockchain technologies: Fu-

ture of transaction processing and smart contracts on

the internet of money. In Banking Beyond Banks and

Money, pages 239–278. Springer.

Prusty, N. (2017). Building Blockchain Projects. Packt Pub-

lishing Ltd.

Prybila, C., Schulte, S., Hochreiner, C., and Weber, I.

(2017). Runtime Veriﬁcation for Business Processes

Utilizing the Bitcoin Blockchain. Future Genera-

tion Computer Systems, page S0167739X1731837X.

arXiv: 1706.04404.

Rademakers, T. (2012). Activiti in Action: Executable busi-

ness processes in BPMN 2.0. Manning Publications

Co.

Smirnov, A., Sheremetov, L., and Teslya, N. (2019). Fuzzy

cooperative games usage in smart contracts for dy-

namic robot coalition formation: Approach and use

case description.

Szabo, N. (1994). Smart contracts. Unpublished

manuscript.

van Der Aalst, W. M. (2009). Workﬂow patterns. Encyclo-

pedia of Database Systems, pages 3557–3558.

Van Der Aalst, W. M. and Ter Hofstede, A. H. (2005). Yawl:

yet another workﬂow language. Information systems,

30(4):245–275.

Viriyasitavat, W., Da Xu, L., Bi, Z., and Sapsomboon, A.

(2018). Blockchain-based business process manage-

ment (BPM) framework for service composition in in-

dustry 4.0. Journal of Intelligent Manufacturing.

Weber, I., Xu, X., Riveret, R., Governatori, G., Ponomarev,

A., and Mendling, J. (2016). Untrusted Business Pro-

cess Monitoring and Execution Using Blockchain. In

La Rosa, M., Loos, P., and Pastor, O., editors, Busi-

ness Process Management, volume 9850, pages 329–

347. Springer International Publishing, Cham.

Exploring Blockchain Technology to Improve Multi-party Relationship in Business Process Management Systems

825