Towards Business Process Model Extension with Cost Perspective

Based on Process Mining

Petri Net Model Case

Dhafer Thabet, Sonia Ayachi Ghannouchi and Henda Hajjami Ben Ghézala

RIADI Laboratory, National School of Computer Sciences, Mannouba University, Mannouba, Tunisia

Keywords: Business Process Management, Business Process Improvement, Process Mining, Petri Net Model

Extension, Cost Information.

Abstract: Business process improvement has ever been the major organizations concern to enhance efficiency,

flexibility and competitiveness. Business process management is a contemporary approach which includes

different techniques to support business process improvement along its lifecycle phases. Process mining is a

maturing technology based on event logs analysis giving insight on what is really happening at the

operational level. Process model extension is one of the three process mining types, which provides

different perspectives of the business process. Otherwise, cost perspective is one of the relevant information

needed by organization managers to take the suited process improvement decisions. In this paper, we

propose an approach allowing Petri Net model extension with cost information using process mining

extension technique. Besides, we present the main details about its implementation and the test case results.

1 INTRODUCTION

Efficiency, flexibility and competitiveness have ever

been the main concerns of all organizations (Briol,

2008). In this context, Business Process

Management (BPM) is a contemporary approach

which has been highly considered for its potential of

enhancing organizations business processes

continuously (Adams, et al., 2010; Rosemann and

vom Brocke, 2010). In our work, we consider the

overall goal of supporting organizations decision

makers to enhance their business processes.

Among the different BPM techniques, process

mining is used to analyze business processes based

on event logs recorded by the systems executing

these processes (Adams, et al., 2010; Rosemann and

vom Brocke, 2010) which makes its results relevant

and accurate for managers and decision makers.

Event logs can be used to conduct three types of

process mining (van der Aalst, 2011; IEEE Task

Force on Process Mining, 2012; Rozinat, 2010): (1)

discovery: produces a business process model using

event logs; (2) conformance: an existing process

model is compared with the corresponding event

logs to identify the eventual deviations; (3)

enhancement: includes two sub-types: repair

(improving the model to better reflect reality) and

extension (using information extracted from event

logs to enrich the model). The extension technique

enables process model enhancement with different

perspectives such as the organizational, the case and

the time perspectives.

Otherwise, organizations are always concerned

with reduction of costs incurred during the execution

of their business processes. Management accounting

is the field dealing with how cost and other

information should be used for planning, controlling,

continuous improvement and decision making

(Weygandt, Kimmel and Kieso, 2010; Hansen and

Mowen, 2006). It includes several techniques such

as: Activity-Based Costing/Management (ABC/M)

(Weygandt, Kimmel and Kieso, 2010; Hansen and

Mowen, 2006); Time-Driven ABC (TDABC)

(Kaplan and Anderson, 2003); and Resource

Consumption Accounting (RCA) (RCA Institute,

2008; Clinton and Webber, 2004). The goal of these

techniques is to measure costs incurred during

process execution and to allocate them to the

business process operations.

These cost information would be easier to be

interpreted and more significant and accurate for

decision makers if they are associated to the

335

Thabet D., Ayachi Ghannouchi S. and Hajjami Ben Ghézala H..

Towards Business Process Model Extension with Cost Perspective Based on Process Mining - Petri Net Model Case.

DOI: 10.5220/0004883003350342

In Proceedings of the 16th International Conference on Enterprise Information Systems (ICEIS-2014), pages 335-342

ISBN: 978-989-758-029-1

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

corresponding elements of the business process

model. This leaded us to consider the issue of

business process model extension with cost

information based on process mining extension

technique. As a first step of our work, we considered

to start by dealing with the case of Petri Net models.

In this paper, we propose an approach as well as its

implementation for cost extension of Petri Net

models using the process mining extension

technique.

In the remainder of this paper, we present the

research questions of our work in Section 2. In

Section 3, we give an overview about related works.

Section 4 presents the basic concepts related to Petri

Nets. Section 5 deals with the description of our

proposed approach. Section 6 presents the

implementation of the proposed approach. Finally,

in Section 7, we summarize our contribution and

present our future works.

2 RESEARCH QUESTIONS

On one hand, the process mining extension

technique enriches the business process model with

a wealth of information about the operational level

based on event logs. On the other hand, cost

information related to the business process is crucial

and interesting for decision making support.

Therefore, the main research question of our work is

how to extend the business process model with cost

information using the process mining extension

technique. This leads us to deal with the two

following questions:

 (1) Is it possible to provide cost information from

event logs? If so, how?

 (2) How to extend the business process model with

the obtained cost information?

The question (1) is about the possibility and then

the way for collecting cost information from event

logs. If we assume that it is possible to do so, the

question (2) deals with the manner by which the

business process model could be extended with the

collected cost information. In this paper, as a starting

point in our work, we propose a solution for the first

research question (1) while we start by dealing with

the second research question (2) for business

processes modelled with Petri Nets. Therefore, the

main contribution of this paper is the proposal of an

approach as well as its implementation providing

accurate cost information as a perspective of Petri

Net models using the process mining extension

technique.

3 RELATED WORKS

Nauta (2011) proposed an architecture to support

cost-awareness in process mining and its first

implementation. The architecture contains two parts:

(1) in the management accounting part, the

accountant provides cost drivers and cost functions.

A cost driver defines how cost is associated with a

process and the cost amount incurred per unit of

activity or time. A cost function defines the content

of a cost report. (2) The Business Process

Management System (BPMS) part consists of 3

main steps. (a) In the first step, cost drivers and

functions, the business process model and the

organizational model are taken as inputs to create a

cost model (Nauta, 2011). (b) During the second

step, event logs together with the cost model are

used to calculate cost information that is used to

annotate the initial event log which is in the XES

(eXtensible Event Stream) format. XES is an XML-

based generic format for event logs (Nauta, 2011;

Wynn, Low and Nauta, 2013). A XES event log

consists of an arbitrary number of traces. Each trace

contains an arbitrary number of events (Hverbeek,

2012). The cost annotation is performed, per cost

type, in the final event of each task instance. The

obtained cost annotated event log is also in the XES

format. (c) In the third step, the cost annotated event

log is used to create cost reports (Nauta, 2011).

The context of this work is a current PhD project

named “Cost-aware BPM” started in 2012 (Wynn,

2012). The author proposes a framework to support

management accounting decisions on cost control

for monitoring, predicting and reporting based on

the work of Nauta (2011). The implemented plug-ins

of this framework allow the generation of different

cost reports (Wynn, Low and Nauta, 2013).

4 PETRI NETS

4.1 Definitions

A Petri Net is a directed bipartite graph populated by

three types of objects: places, transitions, and

directed arcs. Directed arcs connect places to

transitions or transitions to places (

Li and Zhou, 2009;

Wang, 2007). Places are graphically represented by

circles and transitions by boxes or bars. A place can

hold tokens denoted by black dots, or a positive

integer representing their number. The distribution

of tokens over the places is called a marking which

corresponds to a state of the modelled system.

ICEIS2014-16thInternationalConferenceonEnterpriseInformationSystems

336

The initial token distribution is called the initial

marking. A Petri Net is formally defined as a 5-tuple

N = (P, T, I, O, M

), where:

 P = {p

, p

, …, p

} is a finite set of places;

 T = {t

, t

, …, t

} is a finite set of transitions,

P ∪ T ≠ ∅, and P ∩ T = ∅;

 I: P × T → N is an input function that defines

directed arcs from places to transitions, where N is

a set of non-negative integers;

 O: T × P → N is an output function that defines

directed arcs from transitions to places;

 M

: P → N is the initial marking.

4.2 Properties

Petri Nets have a number of properties which allow

the system designer to identify the presence or

absence of the application domain specific

functional properties of the considered system. Two

types of properties can be distinguished: behavioural

and structural properties. The behavioural properties

depend on the initial marking of a Petri Net while

the structural ones don’t depend on it. The most

important behavioural properties are : reachability,

boundedness, safeness, liveness, reversibility (

and Zhou, 2009; Wang, 2007).

4.3 Types

There are a number of types of Petri Nets which may

be classified in three levels (Trompedeller, 1995): (1)

Petri Nets characterised by places which can

represent boolean values (examples: Elementary Net

Systems, State Machines); (2) Petri Nets

characterised by places which can represent integer

values (examples: Marked Graphs, Free Choice

Nets); (3) Petri Nets characterised by places which

can represent high-level values (Examples: High-

Level Petri Nets, Coloured Petri Nets).

5 PROPOSED APPROACH

The proposed solution is an approach allowing the

extension of Petri Net models with cost perspective

using process mining extension technique. In the

following, the approach overview and the Petri Net

meta-model cost extension are presented.

5.1 Approach Overview

As shown in Figure 1, the inputs of our proposed

approach are: the Petri Net, that models the business

process, and the event log which was generated by

the BPMS executing this process and then annotated

with cost data by the application of Nauta’s

approach (2011). The main step of our proposed

approach is the extension of the Petri Net with cost

information. This step consists of three main sub-

steps: (1) In the first sub-step, the cost annotated

event log is used to extract cost information. (2)

During the second sub-step, the Petri Net model is

loaded into memory. (3) In the third sub-step, cost

information is computed with respect to the user’s

requirements and is associated to the corresponding

elements of the Petri Net. The output of this step is a

Petri Net extended with cost information. Finally,

the Petri Net is displayed along with the associated

cost perspective.

Figure 1: Overview of our proposed approach.

5.2 Petri Net Meta-Model Extension

with Cost Data Structure

Because of the different types of Petri Nets, several

works focused on the proposal of a standard

specification for Petri Net models to facilitate their

transfer between different tools.

The Petri Net Markup Language (PNML) is a

Extend the Petri Net Model

with Cost Information

Event log annotated

with cost information

(

Nauta’s a

roach

)

Petri Net

model

Extract

cost information

Cost

information

extracted

Compute and associate

cost information

to the corresponding

Petri Net model elements

Petri Net model

extended with cost perspective

Display

the cost extended Petri Net model

Cost extended Petri Net model

graphically displayed

Load

the Petri Net model

Petri Net model

loade

Input/Output

Step/Sub-Step

Legend

TowardsBusinessProcessModelExtensionwithCostPerspectiveBasedonProcessMining-PetriNetModelCase

337

Figure 2: PNML core model extended with Cost data structure.

proposal of an XML-based interchange format for

Petri Nets and has been defined as an international

standard (ISO/IEC 15909 series) which defines a

meta-model of four packages: PNML core model,

Place/Transition Nets, Symmetric Nets and High-

level Petri Net Graphs. The last three packages are

related to specific versions of Petri Nets while the

first one can represent any kind of Petri Net

(Billington, et al., 2003; LIP6, 2012; 2013; Hillah, et

al., 2009). Therefore, in our work, we consider the

PNML core model package as the Petri Net meta-

model. Figure 2 shows the UML (Unified Modeling

Language) class diagram representing the PNML

core model extended with the cost data structure. A

document that meets the PNML core model is a Petri

Net document which contains one or more Petri

Nets. Each Petri Net has a name and contains one or

more pages. Each page is an object that consists of

objects. Each object has a unique identifier, a name

and may have graphical information. Besides, an

object may be a node or an arc. An arc must connect

nodes of the same page. A node can be whether a

transition node or a place node. A transition node is

whether a transition or a reference transition.

Similarly, a place node is whether a place or a

reference place. Reference transitions and reference

places are used to allow the connection of nodes that

belong to different pages, by an arc. A Petri Net, its

pages and its objects can have labels which may be

annotations or attributes. For some tools, specific

information might be stored for each Petri Net,

object or label (Hillah, et al., 2009).

The cost data structure consists of the two grey

colored classes (CostAnnotation and Cost classes)

and their corresponding relationships as shown in

Figure 2. The Cost class represents the cost concept

which is described by: computation mode, type,

amount (value) and currency. The computation

mode defines how the cost amount is calculated

(average, maximum or minimum). The cost type is

one of the types defined in the cost annotated event

log: fixed cost, labor cost and overhead cost.

Besides, we added a fourth cost type: the total cost

(sum of all the other cost types values). The cost

annotation is a kind of annotation as it represents

information textually displayed. Furthermore, a cost

annotation consists of a set of costs and is associated

to a transition as costs are attributed to tasks in the

cost annotated event log. Moreover, the cost

information of a reference transition is that

associated to its corresponding transition.

6 APPROACH

IMPLEMENTATION

In order to test our approach, we implemented a

prototype allowing the extension of a Petri Net

model with cost information extracted from the

corresponding cost annotated event log.

Implementation and test details are presented in the

next sub-sections.

6.1 Tool Architecture Overview

As shown in the left side column of Table 1, the

required inputs for our tool are the following: (1) the

Petri Net model file (PNML format) which meets

the PNML core model structure; and (2) the

PetriNet

-id

-type

PetriNetDoc

1..*

Page

1..*

Object

#id

Na m e

ToolInfo

-tool

-version

Label

Annotation

Attribute

Graphics

No d e

Arc

PlaceNode

Place RefPlace

TransitionNode

Transition

RefTransition

0..1

+target

+source

Cost

-computation_mode

-cost_type

-amount

-currency

0..1

CostAnnotation

ICEIS2014-16thInternationalConferenceonEnterpriseInformationSystems

338

corresponding cost annotated event log (XES

format) which is produced with respect to the cost

extended XES meta-model (Nauta, 2011; Wynn,

Low and Nauta, 2013). The central part of Table 1

illustrates the internal structure of the implemented

tool which is modelled by an UML component

diagram. The inputs are imported using graphical

user interfaces (GUI package). The extraction of

cost information from the cost annotated event log is

performed using the cost extraction package that

imports the OpenXES (1.9 version) library which is

a reference implementation of the XES standard for

storing and managing event log data (Hverbeek,

2012). Besides, the Petri Net model is loaded using

the cost extended PNML core model package which

uses the PNML framework that is a prototype

implementation of the international standard on Petri

Nets (LIP6, 2012; 2013). In addition to the

orchestration of these functionalities, the other roles

of the main package are the handling of the extracted

cost data, their association to the corresponding

loaded Petri Net elements and the display of the cost

extended Petri Net model using the GUI package.

Then, the output produced by the tool is a Petri Net

model extended with a cost perspective graphically

displayed.

6.2 Cost Extraction and Extension

Algorithms

Figure 3 presents the algorithm of cost information

extraction from cost annotated event log. When

starting the parse of the cost annotated event log, a

log cost object (lco) is initialized. For each trace in

the event log, a trace cost object (trco) is defined.

Each time an end event is found in the current trace,

a task instance cost object (tico) is initialized. For

each cost type element found in the current end

event, the related cost data is extracted and used to

create an elementary cost object (eco) which will be

added to the current tico. Cost objects related to task

instances of one same task in one same trace are

grouped within a task cost object (taco). Once the

current trace is fully parsed, the current trco is added

to the current lco.

Once the cost data is extracted from the cost

annotated event log and the Petri Net model is

loaded with respect to the PNML core model, the

model extension sub-step is performed. Figure 4

shows the Petri Net model cost extension algorithm.

As the cost extracted data is related to tasks in the

cost annotated event log, we consider simply the

extension of transitions contained in each page of

the Petri Net. A cost annotation object (cao) is

initialized each time a transition is found in the

current page. The cao will contain a list of cost

objects. For each combination of the computation

modes and the cost types, selected by the user, a cost

object is created with the corresponding data

(computation mode, type, amount and currency) and

then it is added to the current cao. After that, the cao

is added in turn to the current transition object. Thus,

these instructions are executed for each transition of

each page of the Petri Net.

Table 1: General architecture of the Petri Net model cost extension tool.

Inputs Tool Architecture (UML Component Diagram) Output

XES meta-model

with cost extension

(.xesext)

Cost annotated

event log

(.xes)

Petri Net model

(.pnml)

PNML core model

(.rng)

Cost extended

Petri Net model

graphically

displayed

Input/Output

Legend

Data Structure

is instance of

TowardsBusinessProcessModelExtensionwithCostPerspectiveBasedonProcessMining-PetriNetModelCase

339

Figure 3: Algorithm of the cost information extraction.

Figure 4: Algorithm of the Petri Net model cost extension.

6.3 Test Example

In order to test our approach, we considered the case

of a simple phone repair process. Figure 5 shows the

Petri Net model representing the considered business

process. The process begins by the registration of the

broken phone and then it is analyzed to determine

the defect type. Depending on the severity of the

defect, a simple repair or a complex repair is carried

out. Then, the phone is tested to check whether it is

fixed. If so, the repair details are archived and it is

returned to the customer. If it is not fixed, the repair

and then the test are restarted again. If the phone is

still broken after the fifth repair test, the repair

details are archived and the phone is returned to the

customer. Otherwise, the customer is informed about

the defect type after the defect analysis and before

archiving the repair details (Nauta, 2011; Process

Mining Group, 2013). This process example was

already used as a test case example in Nauta’s

approach. We used the produced cost annotated

event log together with its corresponding Petri Net

model as inputs in our tool test phase.

Figure 5: Petri Net model of the simple telephone repair

process (Nauta, 2011, p.14).

6.4 Test Results

After launching the Petri Net model cost extension

For each page in the Petri Net

For each transition in the page

Initialize a cost annotation object

For each computation mode selected by the user

For each cost type selected by the user

Create and Add corresponding cost object to the cost annotation object

End For

Add the cost annotation object to the current transition object

End For

Initialize a log cost object

For each trace in the event log

Initialize a trace cost object

For each end event in the trace

Initialize a task instance cost object

For each cost type in the end event

Create an elemental cost object using the extracted cost information

Add the elemental cost object to the task instance cost object

End For

If the corresponding task cost object exists in the trace cost object Then

Add the task instance cost object to the corresponding task cost object

Else

Create the corresponding task cost object

Add the task instance cost object to the created task cost object

Add the created task cost object to the trace cost object

End If

End For

Add the trace cost object to the log cost object

End For

ICEIS2014-16thInternationalConferenceonEnterpriseInformationSystems

340

tool, the user selects the cost extension parameters

(Figure 6) including the cost computation mode(s),

the cost type(s), the PNML file (Petri Net model)

and the XES file (cost annotated event log).

Figure 6: Selection of the cost extension parameters.

The user selects one or more of the provided

computation modes: average, maximum and

minimum. Figure 6 shows that the user selected to

compute the average cost. The user also selects one

or more of the provided cost types: fixed, labor,

overhead and total. As shown in Figure 6, the user

selected to get all the provided types of costs. The

Petri Net model and the corresponding cost

annotated event log files are selected by providing

their absolute path. In Figure 6, the selected PNML

and XES files correspond to the simple phone repair

process example. Once the cost extension

parameters are validated by the user, the cost

extended Petri Net is displayed on the main tool

frame (Figure 7). The cost information is displayed

for each transition using tooltips appearing when the

user hovers the mouse over one transition which will

be automatically highlighted. The cost information

related to a transition is described using a table

which entries represent the different combinations

between the computation mode(s) and the cost

type(s) selected by the user. Each entry presents the

corresponding cost value and currency. For instance,

as the user selected to get an average value for each

cost type (Figure 6), each transition of the selected

Petri Net model will be extended with a table

containing an average value of each selected cost

type. Figure 7 shows the cost information related to

the “Analyze Defect” transition which includes an

average total cost of 17.38 AUD (Australian Dollar),

an average fixed cost of 8.20 AUD, an average labor

cost of 3.10 AUD and an average overhead cost of

6.07 AUD.

7 CONCLUSION AND FUTURE

WORKS

In this paper, we proposed an approach for Petri

Nets extension with cost information using process

mining extension technique. Besides, we proposed a

cost extension of the general Petri Net meta-model

(PNML core model) to store the cost information

extracted from the cost annotated event log produced

by Nauta’s approach (2011). Furthermore, we

described the implementation details of our Petri Net

cost extension tool which was tested for the case of a

simple phone repair process.

For this moment, the implemented prototype

allows decision makers to have detailed cost

information directly related to the Petri Net which

helps them better understand their business process

Figure 7: Displaying the cost information related to the “Analyze Defect” transition.

TowardsBusinessProcessModelExtensionwithCostPerspectiveBasedonProcessMining-PetriNetModelCase

341

from an accurate cost point of view which in turn,

may lead them to detect the eventual cost problems

in their business process. Besides, detection of cost

problems in the business process is a mandatory

phase to be able to take the suited actions for

business process enhancement in terms of cost.

Therefore, the main current contribution of our work

is the assistance of decision makers in the first phase

of business process enhancement in terms of cost.

Currently, we are planning to conduct surveys

with management accounting and business process

management experts to improve and deepen the cost

information handling (structuring, analysis and

presentation) in order to make it more suitable and

useful for decision making. In future works, we

consider the generalization of the proposed approach

to support cost extension of any business process

model notation (not only Petri Nets). Finally, real

world case studies will be carried out in order to

evaluate our approach and tool performances.

REFERENCES

Adams, M., Russell, N., ter Hofstede, A. H. M. and van

der Aalst, W. M. P., 2010. Modern Business Process

Automation, Springer. Heidelberg.

Billington, J., Christensen, S., van Hee, K., Kindler, E.,

Kummer, O., Petrucci, L., Post, R., Stehno, C. and

Weber, M., 2003. The Petri Net Markup Language:

Concepts, Technology, and Tools. In ICATPN’03,

24th International Conference on Applications and

Theory of Petri Nets, Springer.

Briol, P., 2008. Ingénierie des Processus Métiers: de

l’élaboration à l’exploitation, Briol Patrice.

Clinton, D. and Webber, S. A., 2004. RCA at Clopay:

Here’s innovation in management accounting with

Resource Consumption Accounting. In Strategic

Finance.

Hansen, D. R. and Mowen, M. M., 2006. Cost

Management: Accounting and Control, Thomson

South-Western. United States of America.

Hillah, L. M., Kindler, E., Kordon, F. and Petrucci, L.,

2009. A primer on the Petri Net Markup Language and

ISO/IEC 15909-2*. In CPN’09 10th International

workshop on Practical Use of Colored Petri Nets and

the CPN Tools. Denmark.

Hverbeek, 2012. Extensible Event Stream. [online]

Available at: <http://www.xes-standard.org>

[Accessed June 2013].

IEEE Task Force on Process Mining, 2012. Process

Mining Manifesto. In BPM 2011 Workshops. Springer.

Kaplan, R.S. and Anderson, S.R., 2003. Time-Driven

Activity-Based Costing. In HBS Working Paper.

Li, Z. and Zhou M., 2009. Petri Nets. In: Deadlock

Resolution in Automated Manufacturing Systems.

Ch.2. Springer. (online) Available at: Springer website

<http://www.springer.com/computer/swe/book/978-1-

84882-243-6> (Accessed July 2013).

LIP6 (Laboratoire d’Informatique de Paris 6), 2012.

PNML Framework download site. (online) Available

at: <http://pnml.lip6.fr> (Accessed July 2013).

LIP6 (Laboratoire d’Informatique de Paris 6), 2013.

PNML reference site. (online) Available at :

<http://www.pnml.org> (Accessed July 2013).

Nauta, W.E., 2011. Towards Cost-Awareness in Process

Mining. Master’s thesis. Eindhoven. (online)

Available at: <http://alexandria.tue.nl/extra1/

afstversl/wsk-i/nauta2011.pdf> (Accessed April 2012).

Process Mining Group, 2013. Process Mining: Research,

Tools and Application. (online) Available at:

<http://www.processmining.org> (Accessed June

2013).

RCA Institute, 2008. Resource Consumption Accounting

Institute. (online) Available at: <http://www.

rcainstitute.org> (Accessed April 2012).

Rosemann, M. and vom Brocke, J., 2010. Handbook on

Business Process Management 1, Springer.

Heidelberg.

Rozinat, A., 2010. Process mining: Conformance and

Extension, PhD Thesis, University Press Facilities.

Eindhoven.

Trompedeller, M., 1995. Petri Nets Classification. (online)

Available at: <http://www.informatik.uni-

hamburg.de/TGI/PetriNets/classification/> (Accessed

July 2013).

van der Aalst, W. M. P., 2011. Process Mining:

Discovery, Conformance and Enhancement of

Business Processes, Springer-Verlag. Berlin.

Wang, J., 2007. Petri Nets for Dynamic Event-Driven

System Modeling. In: P. Fishwick, ed. 2007.

Handbook of Dynamic System Modeling. CRC Press.

[online] Available at: Monmouth University website <

http://bluehawk.monmouth.edu/~jwang/Petri%20Nets

%20--%20Introduction.pdf> (Accessed July 2013).

Weygandt, J. J., Kimmel, P. D. and Kieso, D. E., 2010.

Managerial Accounting Tools for Business Decision

Making, John Wiley & Sons Inc. United States of

America.

Wynn, M., 2012, Dr. Moe Thandar Kyaw Wynn’s

homepage. (online) Available at: <http://www.

yawlfoundation.org/moe/index.html> (Accessed May

2012).

Wynn, M. T., Low, W. Z. and Nauta, W., 2013. A

Framework for Cost-Aware Process Management:

Generation of Accurate and Timely Management

Accounting Cost Reports. In CRPIT’13, Conferences

in Research and Practice in Information Technology.

Australian Computer Society Inc.

ICEIS2014-16thInternationalConferenceonEnterpriseInformationSystems

342