Modelling Workflow with Petri Nets

CA4 BPM PetriNets 1

 Workflow Management Issues
Georgakopoulos,Hornick, Sheth

Workflow
Process Workflow Implementation =workflow
specification application

Executable
Business Process
Workflow model application code
Modelling/
& specification Enactment
Reengineering
language Service/ Run-
(BPM/R)
time Support
2
 Workflows & Petri Nets (PNs)
WFMS give an explicit representation of the BP
logic thus allowing for computerized support
PNs are an established tool for modelling &
analyzing business processes:
Can be used as a design language for the specification of
complex WFs
PN theory provides for powerful analysis techniques for
verifying the correctness of WF procedures.
PN primarily used to study dynamic concurrent
behaviour of n/w-based systems with a discrete
flow.
CA4 BPM PetriNets 3
 Workflows & Petri Nets (cont'd)

Workflows are case-based, i.e., every piece of

work is executed for a specific case.
Case: the subject of operation in a business process
execution. E.g. mortgage application, hospital admission,
insurance claim, tax declaration, order, request for
information...
A workflow process is designed to handle similar
cases. Cases are handled by executing tasks in a
specific order.

CA4 BPM PetriNets 4

A three-dimensional view of a WF

5
(W.M.P. van der Aalst)
 Basics of Petri Nets

●
Petri nets comprise two types of nodes: places and
transitions. An arc exists only from a place to a
transition or from a transition to a place.
●
A place may have zero or more tokens.
●
Graphically, places, transitions, arcs, and tokens are
represented respectively by: circles, bars, arrows,
and dots.

p1 t1 p2

CA4 BPM PetriNets 6

 Dynamic modelling with Petri nets
Transitions are the active components.
often represent an event, an operation, a transformation or a
transportation.

Places are passive.

usually represents a medium, a buffer, a geographical location, a
state a phase or a condition.
depends on how the token is place is interpreted

Tokens often indicate objects.

can play a role as physical object, e.g. a product/person;
an info object, e.g. a message;
an indicator of state a process is in or state of an object;
an indicator of a condition, i.e. the presence of a token indicates
7
whether a certain condition is fulfilled.
Object Life Cycle (OLC) with Petri Nets
A Petri net attaches to a life cycle of objects of a
class
States correspond to places
Initial state: state with token, there is only one
initial state in an OLC
Transitions correspond to events, conditions
(verify a condition) or processes (or atomic
process: method) that changes object state
Tokens represent objects in this class 8
 Basics of Petri Nets (cont'd)

P1
Place T1
P2
or Transition T2

Arc P3 P4
Token T3
P5
9
 Example – claims process

transition place
token

Pay

Claim Ready
Record
Under
consideration
Send letter

State:
(3,0,0)
(2,1,0)
(2,0,1)
(1,1,1)
(1,0,2)

10
 Basics of Petri Nets (cont'd)
●
Below is an example Petri net with two places and
one transaction.
●
Transition node is ready to fire if & only if there
is at least one token at each of its input places

t1 p1 p2
state transition of form (1, 0) -> (0, 1)
p1 : input place p2: output place

CA4 BPM PetriNets 11

 Formal Notation of Petri Nets
A bipartite graph, PN=(P, T, I, O)
P: finite set of places
T: finite set of transitions
I: (P*T)→ N, I(p,t)=n, if n>0, p∈P, t ∈T, then p is an input place
of t; n is an input multiplicity (weight) for each input arc (p,t)
O: (T*P)→ N, O(t,p)=m, if m>0, p∈P, t ∈T, then p is an output
place of t; m is an output multiplicity(weight) for each output
arc (t,p)
By default, the weight of an arc is equal 1, otherwise it will be noted.
The input multiplicity of an arc between an input place and a transition
determines how many tokens have to be present in the place so that the
transition is enabled
12
Formal Notation of Petri Nets (cont'd)

A state of a Petri net is a function s: P → N, assigning to each

place p ∈P a number of tokens at this place. A state space of a
Petri net is a set of all s(p), p∈P. (E.g. state space is (2,1, 0, 0, 0))

A transition t is enabled, t ∈ T in state s: P → N, if there are

enough tokens present in each of the input places of t, i.e. if and
only if ∀p∈P, s(p) ≥ I(p,t)

A transition t can fire in a state s whenever it is enabled in this

state. When it fires, it consumes I(p,t) tokens from each input
place p and produces O(t,q) tokens in each output place q.
If t fires in state s, this leads to a new state s’ where ∀p∈P,
s’(p)=s(p) – I(p,t) + O(t,p)

13
Properties of Petri Nets
●
Sequential Execution
Transition t2 can fire only after
the firing of t1. This impose the
precedence of constraints "t2 p1 t1 p2 t2 p3
after t1."
●
Synchronization
Transition t1 will be enabled
only when a token there are at
least one token at each of its
input places.
t1
●
Merging
Happens when tokens from
several places arrive for
service at the same transition.
CA4 BPM PetriNets 14
Properties of Petri Nets
(contd)
●
Concurrency
t1 and t2 are concurrent.
With this property, Petri nets t1
can model systems of
t2
distributed control with
multiple processes executing
concurrently in time.

CA4 BPM PetriNets 15

Properties of Petri Nets
(contd)
●
Conflict
t1 and t2 are both ready to fire
but the firing of one leads to t1
the disabling of the other
transitions. t2

t1

t2

CA4 BPM PetriNets 16

 Properties of Petri Nets (contd)

●
Conflict - (contd)

●
the resulting conflict may be resolved in a
purely non-deterministic way or in a
probabilistic way, by assigning appropriate
probabilities to the conflicting transitions. e.g:

t1 t2

CA4 BPM PetriNets 17

t3 t4
 Example: Patients & a Specialist

free wait

start end start

change

busy change documenting inside treated

Tokens : Specialist Tokens : Patient

(W.M.P. van der Aalst)
18
 Example: Patients & a Specialist
(cont'd)
The process of a specialist treating patients:

end

busy documenting
free

wait
start inside chang treated
e
If a specialist always treats two patients at the same time ?
19
Example: Patients & a Specialist
(cont'd)
end

busy documenting
free

2 2 2 2
wait
start inside change treated

20
 Example: In a Restaurant
Waiter
Customer 1 free Customer 2

Take Take
order order

wait Order wait

taken

eating eating
Tell
kitchen Serve food
Serve food

CA4 BPM PetriNets 21

 Example: In a Restaurant (cont'd)
Two Scenarios

Scenario 1:
Waiter takes order from customer 1; serves customer 1;
takes order from customer 2; serves customer 2.
Scenario 2:
Waiter takes order from customer 1; takes order from
customer 2; serves customer 2; serves customer 1.

CA4 BPM PetriNets 22

Example: In a Restaurant (Scenario 1)
Waiter
Customer 1 free Customer 2

Take Take
order order

wait Order wait

taken

eating eating
Tell
kitchen Serve food
Serve food

CA4 BPM PetriNets 23

Example: In a Restaurant (Scenario 2)
Waiter
Customer 1 free Customer 2

Take Take
order order

wait Order wait

taken

eating eating
Tell
kitchen Serve food
Serve food

CA4 BPM PetriNets 24

 Example: Vending Machine

Take 15c bar

Deposit 10c
15c
5c
Deposit 5c

Deposit
0c 5c Deposit
5c Deposit
5c

Deposit 10c
10c 20c
Deposit 10c
Take 20c bar

CA4 BPM PetriNets 25

 Example: Vending Machine (3
Scenarios)

Scenario 1:
Deposit 5c, deposit 5c, deposit 5c, deposit 5c, take
20c snack bar.
Scenario 2:
Deposit 10c, deposit 5c, take 15c snack bar.
Scenario 3:
Deposit 5c, deposit 10c, deposit 5c, take 20c snack
bar.

CA4 BPM PetriNets 26

Example: Vending Machine (Token Games)

Take 15c bar

Deposit 10c
15c
5c
Deposit 5c

Deposit Deposit
0c 5c Deposit 5c
5c

Deposit 10c
10c 20c
Deposit 10c
Take 20c bar

CA4 BPM PetriNets 27

 Example: Insurance complaint process

To manage different cases, two solutions:

1. Token is added a value (case identifier or colour) for distinguish different cases
28
2. Each case corresponds to a unique instance of the Petri nets
 Petri Nets over Time

●
1962 - Carl Petri originally proposed Petri Nets
without any notion of time. Concept of time was
intentionally avoided because addition of time
restricts the behavior of the net.
●
1970s ~ - Addition of time has been discussed in
order to analyze the performance of modelled
system.
●
Many properties are still undecided for Petri nets
extended with data and time.

CA4 BPM PetriNets 29

 References
http://www.wfmc.org/standards/model.htm
“The Application of Petri Nets to Workflow Management”,
W. van der Aalst, J. Circuits, Systems & Computers, Vol.
8(1) (1998), pp. 21-66
“Design and control of workflow process”, Hajo A.Reijers,
LNCS Springer, 2003
“Coupling Object-Oriented and Workflow Modelling in
Business and Information Process Reengineering”,
Gregory N. Mentzas, IOS Press, 1999
“Workflow Management Coalition Terminology and
Glossary” (WFMC-TC-1011). Technical report, Workflow
Management Coalition, Brussels, 1996.
CA4 BPM PetriNets 30