Process Mining - listed

lifeguardknowledgeInternet και Εφαρμογές Web

30 Ιουλ 2012 (πριν από 5 χρόνια και 1 μήνα)

276 εμφανίσεις

Digitalizing the Dental Oral Implant
Treatment Workflow: Quantification of the
Impact on Precision and Efficiency


Ronny Mans
,

Eindhoven University of Technology, The Netherlands

Hajo Reijers
, Eindhoven University of Technology, The Netherlands

Daniel
Wismeijer
,

Academic Centre for Dentistry Amsterdam, The Netherlands

Michiel van Genuchten
,
Open Digital Dentistry, Switzerland

Abstract

JMIS journal.

Not more than 150 words…

Document should not be more than 40 pages…

Keywords and phrases: Business Process

Simulation, Discrete Event Simulation, Process Mining,
Digital Dentistry.

I
ntroduction

Related Work

Process Mining

Process mining is applicable to a wide range of systems. These systems may be pure
information systems (e.g., ERP systems) or systems where
the hardware plays a more
prominent role (e.g., embedded systems). The only requirement is that the system produces
event logs, thus recording (parts of) the actual behavior.


An interesting class of information systems that produce event logs are the so
-
c
alled Process
-
Awa
re Infor
mation Systems (PAISs)
(Dumas, 2005)
. Examp
les are classical WfMSs (e.g.
Staffware), ERP systems (e.g. SAP), case handling systems (e.g. FLOWer), PDM systems (e.g.
Windchill), CRM systems (e.g. Micro
soft Dynamics CRM), middleware (e.g., IBM's WebSphere),
hospital information systems (e.g., Chipsoft), etc. These systems provide very detailed
information about the activities that have been executed.

world
(
software
)
system
(
process
)
model
event
logs
supports
/
controls
business processes
people
machines
components
organizations
models
analyzes
specifies
configures
implements
analyzes
records
events
(
messages
,
transactions
,
etc
.)
discovery
conformance
enhancement

Figure
1
: Three types of process mining: (1) discovery, (2) conformance, and (3) enhancement.

However, not only PAISs are recording events. Also, in dentistry a wide variety of systems
record events. For example, in a dental practice, a pra
ctice management system is used which
records for each patient the services that have been delivered but also the appointments that
have taken place. In a dental lab an order tracking system may be used which records all steps
that have been taken in order

to deliver a dental product together with the time of their
completion. However, although in dentistry typically multiple business entities (e.g. a dentist
and a dental lab) are collaborating in order to come to a final product, typically the systems
used

are limited to the work practices of one business entity only, e.g. a connection between
multiple systems is typically lacking.


The goal of process mining is to extract information (e.g., process models) from these logs, i.e.,
process mining describes a
family of a
-
posteriori analysis techniques exploiting the information
recorded in the event logs
(Aalst, 2011)
. Typically, these approaches assume that it is possible
to sequentially record events such that each event

refers to an activity (i.e., a well
-
defined step
in the process) and is related to a particular case (i.e., a process instance). Furthermore, some
mining techniques use additional information such as the performer or originator of the event
(i.e., the per
son/resource executing or initiating the activity), the timestamp of the event, or
data elements recorded with the event (e.g., the size of an order).



Process mining addresses the problem that most ‘process/system owners’ have limited
information about w
hat is actually happening. In practice, there is often a significant gap
between what is prescribed or supposed to happen, and what actually happens. Only a concise
assessment of reality, which process mining strives to deliver, can help in verifying proce
ss
models, and ultimately be used in system or process redesign efforts.
The idea of process
mining is to discover, monitor and improve real processes

(i.e., not assumed processes)
by
extracting knowledge from event logs
. We consider three basic types of p
rocess mining (as
shown in Figure 1): (1)
discovery
, (2)
conformance
, and (3)
extension
.





Discovery:

The first type of process mining is discovery, i.e., deriving information from some
event log without using an a priori model. Based on an event log vario
us types of models
may be discovered, e.g., process models, business rules, organizational models, etc. For
example, many techniques have been developed regarding the discovery of the control
-
flow
perspective, e.g. expressed in terms of Petri nets



(Aalst, 2003)
,

(Aalst, 2004)
, Heuristics
nets

(Aalst, 2007)
,

(Medeiros, 2007)
,

(Weijte
rs, 2003)
, or Event
-
driven Process Chains

(EPCs)

(Dongen, 2004)
. However, process mining is not limited to control
-
flow. Recent
process mining techniques are more and more focusing on other perspectives, e.g.,
the
organizational perspective

(Aalst, 2005)
, the performance perspective

(Dongen, 2008)
,

and
the data perspective

(Rozinat, 2008)
,

(Rozinat, 2009)
,

(Song, 2008)



Conformance:

Here the event log is used to check if reality conforms to a model
\
cite{anneconformanceIS}. For example, there may be a guideline indicating that the time
betwe
en placing and exposing an implant must be at least four weeks while in reality this
does not happen. Conformance checking may be used to detect deviations, to locate and
explain these deviations, and to measure the severity of these deviations.



Extension:

There is an a
-
priori model. This model is extended with a new aspect or
perspective, i.e., the goal is not to check conformance but to enrich the model with the data
in the event log. An example is the extension of a process model with performance data,
i
.e., some a
-
priori process model is used onto which bottlenecks are projected.


Note that there is a clear difference between process mining and the Business Intelligence (BI)
tools as they are used today. BI tools focus on performance indicators such as t
he number of
final restorations placed, the length of waiting lists, and the success rate of surg
ery. As such, BI
tools do
not

show the
end
-
to
-
end process

and
cannot zoom into selected parts

of this process
\
cite{process
-
mining
-
book
-
2011}. Con
versely, proc
ess mining looks ‘inside the process’

at
different abstraction levels.


ProM has become the de facto standard for process
mining. The ProM framework
(www.processmining.org) is a ‘
plug
-
able


environment for process mining using MXML, SA
-
MXML, or XES as inpu
t format. ProM 5.2 was released in 2009. ProM 6 (released in November
2010) provides a completely new architecture and user
-
interface to overcome some of the
limitation of earlier versions of ProM. In this paper, we use both ProM 5.2 and ProM 6.

Data and A
nalysis


Placement of a Prosthesis

start
p
13
end
p
9
p
5
p
2
p
3
p
10
p
1
p
12
p
4
p
6
p
7
p
8
p
14
p
11
D
:
fit prosthesis
D
:
place
prosthesis
D
:
place
implants
D
:
intake
D
:
fit prosthesis
(
front
)
D
:
checklist
D
:
fit prosthesis
(
molars
)
L
:
prepare
registration bite
L
:
prepare for fit
L
:
finish
prosthesis
L
:
prepare for fit
D
:
adjust prosthesis
+
check
-
up
D
:
adjust
prosthesis
D
:
consultation
D
:
check
-
up
D
:
discuss
treatment plan
D
:
evaluation
D
:
consultation
L
:
make
i
ndividual
impression trays
D
:
first
impressions
D
:
register

bite
D
:
individual
impressions
0
.
09
0
.
19
0
.
13
0
.
05
0
.
54
0
.
07
0
.
14
0
.
02
0
.
41
0
.
88
0
.
12
0
.
45
0
.
55
0
.
60
0
.
40

Figure
2
: Workflow regarding the placement of a prosthesis

in the AS
-
IS situation. The prefix of
a task indicates whether the step is performed by a dentist (D) or the dental lab (L).

In case a
place is colored pink, this indicates that there is a high waiting time in the place whereas a blue
and a yellow color i
ndicate that the waiting time is respectively low and medium.

Furthermore,
in case a place has multiple outgoing arcs, for each arc the probability is shown

that it will

be
followed.




Placement of a Crown

p
1
start
p
7
p
5
p
4
p
3
p
2
p
9
p
6
D
:
expose
implants
D
:
place crown
D
:
make
impressions
D
:
place
implants
D
:
check
-
up
D
:
check
D
:
checklist
+
study models
D
:
extraction
D
:
checklist
D
:
intake
D
/
H
:
hygiene
L
:
make individual
impression trays
D
:
consul
-
tation
D
:
discuss
treatment plan
D
/
H
:
hygiene
L
:
make crown
p
8
0
.
05
0
.
21
0
.
10
0
.
18
0
.
09
0
.
09
0
.
05
0
.
23
0
.
75
0
.
25
0
.
11
0
.
51
0
.
38
0
.
92
0
.
08
0
.
11
0
.
75
0
.
16

Figure
3
:
Workflow regarding the placement of a prosthesis in the AS
-
IS situation. The prefix of
a task indicates whether the step is performed by a dentist (D)
, a dental hygienist (H),

or the
dental lab (L).

The color of the places indicat
es the amount of waiting time spent in the
respective place.


Digital Design of Crowns and Bridges

p
5
start
p
6
p
7
p
8
p
9
p
11
p
10
p
12
p
13
p
13
p
14
p
4
end
p
2
p
1
Scan Import
Done
Design Done
CAD Engine
Done
Unlocked
Scan Done
Locked by adm
Locked by adm
Unlocked
Locked by adm
Unlocked
Unlocked
Locked by
adm
Unlocked
Unlocked
Locked by adm
Locked by adm
Unlocked
start
end
0
.
95
Creation and
First
i
nitialization
Unlo
cked
0
.
05
0
.
90
0
.
10
0
.
36
0
.
64
0
.
35
0
.
65
0
.
99
0
.
01
0
.
97
0
.
02
0
.
01
0
.
8
0
.
2
0
.
89
0
.
11

Figure
4
: Workflow for designing crowns and bridges using the Dental Wings CAD/CAM
software.

Intra
-
Oral

Scanning


Figure
5
: Bar chart depicting the average time needed for the intra
-
oral scan of one and two
implants.

Guided Surgery


Figure
6
: Bar chart depicting
,

for different numbers of implants,
the aver
age time needed for the
planning of the guided surgery using the CoDiagnostiX software.

0
5
10
15
20
25
30
35
1
2
average time (minutes)

number of implants

1
2
0.00
10.00
20.00
30.00
40.00
50.00
60.00
2
3
4
5
6
average time (minutes)

number of implants

2
3
4
5
6
Simulation

Making the Simulation Model

Validation

Experiments

Experiment 1: Intra
-
Oral Scan of the Teeth and Digital Design of a Crown


Experiment 2: Intra
-
Oral Scan of

the Teeth and Digital Design of a Prosthesis


Experiment 3: Mini Implants, Computer Guided Surgery, and Digital Design of a
Prosthesis


Discussion


Conclusion

REFERENCES

Journals: authors, title, journal (italic), volume and issue, data, inclusive pages

Books, conference proceedings: title (principal words capitalized, title italic), title ends with .
followed by city : , publisher, year

Book chapters / conf proc: authors, title, “In,” names of the editors, “(ed./eds.)”, title of the
book (princip. Words
capitalized, title italic), city, “:”, publisher, year, “pp.” inclusive pages

Bibliography

Process Mining: Discovery, Conformance and Enhancement of Business Processes

[Boek]

/
aut.
Aalst W.M.P. van der.

-

[sl]

: Springer
-
Verlag, Berlin,
2011.

Process
-
Aware Information Systems: Bridging People and Software

[Boek]

/ aut.
Dumas M.,
Aalst, W.M.P. van der, Hofstede, A.H.M. ter.

-

[sl]

: Wiley & Sons, 2005.



FIGURES


Figure
7
: R
esults for the experiments in whi
ch an Intra
-
Oral Scan of the teeth is made and the
final
prosthesis

is made using CAD/CAM techniques. For the total throughput time of the entire
workflow (avgTTCrown), the average total time spent by people in the lab (avgLabCrown), and
the average total
time spent by a dentist (avgDentCrown) the average (avg) and standard
deviation (sd) of 100 runs are shown in the ‘simulation results’ table part.
In addition, in the ‘T
-
TEST’ table part, the result of t
-
tests are shown to determine whether the observed av
erage of
two experiments is statistically significant from zero. For each

experiment, the average for each
performance measure is visualized in the graph.

0
50
100
150
200
250
300
avgTTCrown (days)
avgLabCrown (min)
avgDentCrown (min)

SIMULATION

RESULTS

as
-
is

crown08

crown09

crown10

crown11

crown12

avg

sd

avg

sd

avg

sd

avg

sd

avg

sd

avg

sd

avgTTCrown

(days)

231.1

5.4

192.0

4.9

193.6

5.7

193.3

4.9

195.0

5.2

195.6

4.9

avgLabCrown

(min)

180.7

2.8

85.8

1.5

85.7

1.5

85.6

1.5

85.9

1.5

85.7

1.4

avgDentCrown

(min)

229.9

5.7

238.9

5.9

239.2

5.6

239.1

6.2

238.9

5.7

238.5

6.2

T
-
TEST

as
-
is <
-
> crown10

crown08 <
-
>
crown09

crown09 <
-
> crown10

crown10 <
-
>
crown11

crown11 <
-
>
crown12


t
-
value

P
-
value

t
-
value

P
-
value

t
-
value

P
-
value

t
-
value

P
-
value

t
-
value

P
-
value

avgTTCrown

51.7

0.00 *

-
2.20

0.03 *

0.36

0.72

-
2.32

0.02 *

-
0.78

0.44

avgLabCrown

301.00

0.00 * +

0.55

0.58

0.27

0.79

-
1.50

0.14

1.31

0.19

avgDentCrown

-
10.88

0.00 *

-
0.35

0.72

0.14

0.89

0.17

0.87

0.57

0.57


Figure
8
:
Results for the experiments in which an Intra
-
Oral Scan of the t
eeth is made and the
final crown is made using CAD/CAM techniques. For the total throughput time of the entire
workflow (avgTTCrown), the average total time spent by people in the lab (avg
)

and standard
deviation (sd
) of 100 runs are shown in the ‘simulation results’ table part.
In addition, in the ‘T
-
TEST’ table part, the result of t
-
tests are shown to determine whether the observed average of
two experiments is statistically significant from zero. For each

experimen
t, the average for each
performance measure is visualized in the graph.

0
50
100
150
200
250
300
350
400
450
avgTTProth (days)
avgLabProth (min)
avgDentProth (min)

SIMULATION

RESULTS

as
-
is

proth1_
08

proth1_
09

proth1_
10

proth1_
11

proth1_
12

avg

sd

avg

sd

avg

sd

avg

sd

avg

sd

avg

sd

avgTTProth
(days)

250.0

6.2

224.5

5.6

227.4

5.9

229,7

6.7

232.5

6.2

235.8

5.6

avgLabProth
(min)

377.1

6.3

317.9

5.3

317.7

5.1

317.4

4.6

318.0

4.4

317.6

4.5

avgDentProth
(min)

395.0

6.3

365.4

5.5

365.7

5.6

365.8

5.5

365.7

6.2

365.0

5.3

T
-
TEST

as
-
is <
-
> proth1_10

proth1_
08

<
-
>

proth1_
0
9

proth1_
0
9
<
-
>

proth1_10

proth1_10
<
-
>

proth1_11

proth1_11 <
-
>
proth1_12


t
-
value

P
-
value

t
-
value

P
-
value

t
-
value

P
-
value

t
-
value

P
-
value

t
-
value

P
-
value

avgTTProth

22.37

0.00 *

-
3.50

0.00 *

-
2.57

0.01 *

-
3.10

0.00 *

-
4.03

0.00 *

avgLabProth

76.79

0.00 * +

0.25

0.79

0.46

0.65

-
0.93

0.35

0.67

0.50

avgDentProth

34.94

0.00 *

-
0.29

0.77

-
0.23

0.82

0.18

0.86

0.90

0.37


Figure
9
:
Results for the experiments in which
three mini implants are placed, the implants are
placed using guided surgery, and
he final prosthesis is made
using CAD/CAM techniques. For
the total throughput time of the entire workflow (avgTTCrown), the average total time spent by
people in the lab

(avg
)

and standard deviation (sd) of 100 runs are shown in the ‘simulation
results’ table part.
In addition, in t
he ‘T
-
TEST’ table part, the result of t
-
tests are shown to
determine whether the observed average of two experiments is statistically significant from zero.
For each

experiment, the average for each performance measure is visualized in the graph.


0
50
100
150
200
250
300
350
400
450
avgTTProth (days)
avgLabProth (min)
avgDentProth (min)

SIMULATION

RESULTS

as
-
is

proth2_
08

proth2_
09

proth2_
10

proth2_
11

proth2_
12

avg

sd

avg

sd

avg

sd

avg

sd

avg

sd

avg

sd

avgTTProth
(days)

250.0

6.2

165.9

5.8

169.8

5.8

172.8

5.0

176.1

6.0

177.4

5.5

avgLabProth
(min)

377.1

6.3

349.8

5.1

349.3

4.2

349.4

4.3

350.0

4.5

349.0

4.5

avgDentProth
(min)

395.0

6.3

367.1

2.2

367.3

2.2

367.4

2.0

367.2

2.2

367.3

2.5

T
-
TEST

as
-
is <
-
> proth2_10

proth2_
08

<
-
>

proth2_
0
9

proth2_
0
9
<
-
>

proth2_10

proth2_10
<
-
>

proth2_11

proth2_11 <
-
>
proth2_12


t
-
value

P
-
value

t
-
value

P
-
value

t
-
value

P
-
value

t
-
value

P
-
value

t
-
value

P
-
value

avgTTProth

96.78

0.00 * +

-
4.75

0.00 *

-
3.85

0.00*

-
4.38

0.00 *

-
1.56

0.12

avgLabProth

36.41

0.00 *+

0.65

0.52

-
0.21

0.84

-
1.19

0.24

1.88

0.06

avgDentProth

42.01

0.00 *+

-
0.66

0.51

-
0.14

0.89

0.79

0.43

-
0.50

0.61