ABSTRACT

The demand of current economical planet has brought in focus the need to use resources effectively. Todays production management techniques have placed a greater emphasis on the need to plan the manufacturing schedule effectively, in order to have least cost of manufacturing per part and hence improve the productivity. Petri net is a tool for the study and analysis of manufacturing systems. It is in fact a graphical representation that allows a system to be modeled. Analysis of the Petri net can then divulge important information about the structure and the dynamic behavior of the modeled system and suggest ways to improve the efficiency and productivity. Thus, the development of the theory of Petri net is based on the application of Petri net in productivity management. In this paper we have developed a Petri net model for job shop situation in Fuse Shop at Ordnance Factory, Ambajhari, Nagpur. In this case, we have found a much better job-machine-job scheduling giving a more efficient machine utility and leading to drastic improvement in production efficiency.

1.0 INTRODUCTION
Close your eyes and imagine a company that always produces defect free products within well planned time frames. Or a workshop which utilizes raw

materials with minimum wastage and maximum production rate. Dont these results sound impressive? Of course they do. Such things were difficult to imagine even a decade ago but not anymore. The emergence of operations research and production management tools like Six Sigma, Just in time manufacturing, Group Technology, Flexible Manufacturing Systems and more recently of PETRI NET has provided us with the means to improve productivity within cost effective and economic limits. In this paper a detailed study of Classical Petri nets and their application for productivity improvement has been done. 1.1 What is a Petri Net? A Petri net is a graphical and mathematical modeling tool. It is a formal and graphical appealing language which is appropriate for modelling systems with concurrency. Petri nets has been under development since the beginning of the 60'ies where Carl Adam Petri defined the language. It was the first time a general theory for discrete parallel systems was formulated. The language is a generalisation of automata theory such that the concept of concurrently occurring events can be expressed.

1.2

A Petri Net Structure A Petri net is composed of four parts: a set of places P, a set of transitions

T, an input function I, and an output function O. The input and output function relate the transitions and places. The input function I is mapping from transition tj to collection of places I (tj), known as input places of transition. The output function O maps a transition tj to a collection of places O (tj), known as the output places of transition. The structure of a Petri net is defined by its places, transitions, input function and output function. Its structure is a four-tuple one: C = (P, T, I, O) Where, P is a finite set of n places (n>0) T is finite set of m transitions (m>0) PT= I: T of places. O: T bags of places P

P= {p1, p2, p3, p4 pn} T= {t1, t2, t3, t4 tm}

is input function, a mapping from transitions to bags

is the output function, a mapping from transitions to

1.3

Petri Net Graphs Petri nets can be used as a visual-communication aid similar to flow

charts, block diagrams, and networks. In addition, tokens are used in these nets to simulate the dynamic and concurrent activities of systems. The places and

transitions can be graphically represented in a net formed by circles, bars and dots with arrows (or arcs) connecting the places (circles) to the transitions (bars) and tokens (dots) to represent the flow of resources. Input arcs connect places with transitions, while output arcs start at a transition and end at a place. Multiple inputs to transitions are indicated by multiple arcs from input places to transition and similarly multiple outputs are represented by multiple arcs from transition to output places. A Petri Net graph for the following structure is shown in the figure below (figure 1). C = { P, T, I, O } P = { p 1, p 2, p 3, p 4, p 5 } T= { t1, t2, t3, t4 } I(t1) = {p1} I(t2) = { p2, p3, p5 } I(t3) = {p3} I(t4) = {p4} O(t1) = { p2, p3, p5 } O(t2) = {p5} O(t3) = {p4} O(t4) = { p2, p3 }
p3

t4 p1 p5 t2 p4

t1 t3 p2

Figure 1. A general Petri Net graph. 1.4 Petri Net Markings A marking in a Petri Net is an assignment of tokens to the places of the net. Tokens are assigned to , and can be thought to reside in , the places of the Petri net and their number and position can change during the execution of the process of which the net is made. The tokens are denoted by dots and they represent the flow of resources and jobs in the system.

1.5

States , Incidence and Reachability in Petri Net. The state of a Petri net is defined by its marking. Firing of a transition

represents the change in state of the Petri net. The change in state caused by firing a transition is defined by change in function called the next state function. The next state function is unique for each Petri net and is mathematically derived. The incidence of a Petri net is a matrix which defines the order in which the system operates. The incidence matrix specifies the sequence of operation of the various components of the system and is of vital importance to the graphical formulation of the Petri net structure. Once a Petri net has been drawn , it is important to analyse the net to determine what sort of properties it has in order to determine if the design will be feasible for a particular application. The first step is to form the Reachability

matrix which will indicate all the states of the system. The reachability set is a set of all markings reachable from initial marking by firing zero, one , two or more transitions. 1.6 HISTORY AND DEVELOPMENT OF PETRI NET Petri nets are a promising tool for describing and studying systems. The concept of Petri nets has its origin in Carl Adam Petri's doctoral dissertation Kommunikation mit Automaten ( Communication with Automata) submitted in 1962 to the faculty of Mathematics and Physics at the Technische Universitt Darmstadt, Germany. Subsequently Petris work came to the attention of researchers working on Information System Theory Project at Applied Data Research Inc. which led to the development of much of the early theory, notation and representation of Petri nets. Petris work also came to the attention of researchers at Massachusetts Institute of Technology, Boston and led to its further development and application of Petri net to networks, communication systems, computer architecture, system software and manufacturing systems. The Computation Structures group, under the direction of Professor Jack. B. Dennis has been the source of considerable research and publication on Petri nets. The first workshop on Petri net was held in Paris in 1977 and the first advanced course on Petri net was started in Hamburg, Germany in 1979. A special interest group has been formed in many countries in Europe and United States of America to work on the application and development of Petri Net theory.

2.0 APPLICATION OF PETRI NET

Petri nets are a promising tool for describing and studying systems that are characterized as being concurrent, asynchronous, distributed, parallel, nondeterministic, and/or stochastic. As a graphical tool, Petri nets can be used as a visual-communication aid similar to flow charts, block diagrams, and networks. In addition, tokens are used in these nets to simulate the dynamic and concurrent activities of systems. As a mathematical tool, it is possible to set up state equations, algebraic equations, and other mathematical models governing the behavior of systems. The generality and permissiveness , inherent in Petri nets make it widely applicable which and its use is steadily gaining momentum. A few applications of Petri nets are described underneath. Design of Flexible Manufacturing Systems and performance evaluation. Job shop modeling and tool management in shop floor. Analysis of system states and processes. Modeling of social system and project management. The figure (figure 2.0) in the following page shows the application of Petri net to model a hospital. Let us assume that there are three doctors , one each for Consultancy, Surgery and Dressing. Now three patients with varying requirements arrive at the

hospital. The incidence matrix (table 1.0) shows us the allotment of each patient to the docors with respect to necessity. FACILITIES PATIENTS PATIENT-1 1 (P-1) PATIENT-2 --(P-2) PATIENT-3 1 (P-3) Table 1. Incidence Matrix for Figure 2. 1 1 1 1 --1 CONSULTANCY (CS) SURGERY (SG) DRESSING (DS)

P S 1 2 .. N+ 1

1 1 0 .. 1 Table

2 1 0 .. 1

3 1 0 .. 1

4 1 0 .. 1

5 1 1 .. 1

6 1 1 .. 1

7 0 1 .. 0

8 0 1 .. 0

9 0 0 .. 0

10 11 0 1 .. 0 0 1 .. 0

12 0 0 .. 0

13 0 0 .. 0

14 0 0 .. 0

2. Reachability Matrix for the Petri graph 2.0) net (figure

P4 P1 P2

P5

P3 T1 P6 P7

T2 P8

T3 P9

T4 P11

P10 T5 T6

P12

P13 T7

P14

Figure 2. Petri Net graph for the medical facility. The Reachability matrix shown above (table 2.0 where S denotes states and P denotes places) shows us the various states of the Petri net during its execution. The presence of tokens in places (P1, P2, P3 as shown in fig. 2.0) is denoted by placing 1 and the remaining places are filled with 0. With the change of state (as a transition T1, T2.. takes place) the tokens flow and occupy different places (column 2 in table 2.0) denoting the execution of an activity. Thus with each state the tokens keep on flowing and finally at one point the first state is repeated. This denotes that the designed Petri net is deadlock free and correct.

The Reachability matrix also tells us about the availability of the resources at any instant of time. The idle resources can then be utilized effectively so as to increase the efficiency of the system.

3.0 CASE STUDY

After the detailed analysis of the theory of Petri net, we decided to go for a practical example of modeling a job shop situation at Fuse Shop at Ordnance Factory Ambajhari, Nagpur. The suitability of the job shop was mainly due to the job shop practices and concurrent machine operations performed in it. The following assumptions were made: 1. The operations in the shops occur sequentially and take zero time. 2. The machines are ready for work without any break down or fault. 3. The raw materials are available whenever the process takes place.

Ammunition Fuse shop The Ammunition Fuse shop is engaged in manufacturing various types of ordnance and ammunition fuses. A fuse is an integral part of an explosive and it is responsible for its timely ignition. We have considered the case of manufacturing of B 429 ammunition fuse for 130mm HE bomb shell. The fuse consists of the following parts: Main Body Detent Striker Detonator

The incidence matrix for the manufacturing of these fuse parts along with the machines used for the process is shown in table 3.

SINGLELATHE SPINDLE MACHINE MACHINE (M-1) (M-2)

MULTISPINDLE MACHINE (M-3)

DRILLING MACHINE

MILLING MACHINE

(M-4)

(M-5)

MAIN BODY (J-1) DETENT --(J-2) STRIKER --(J-3) DETONATO R (J-4) Table 3. Incidence Matrix for figure 3. The Petri net model for the above situation, by considering sequential execution is shown in figure 3.The legends for the fig. 3 are detailed underneath (table 4.0). It has six transitions and twelve places. ------1 1 --1 ----1 ------1 ----1 ---

Transitions t1 = Job 1 is to be processed by machine 1. t2 = Job 2 is to be processed by

Places p1 = Job 1 is idle and ready to be processed by machine 1. p2 = Machine 1 is idle and ready to machine job 1.

machine 2. t3 = Job 3 is to be processed by machine 3. t4 = Job 1 is to be processed by machine 4. t5 = Job 4 is to be processed by machine 4. t6 = Job 4 is to be processed by machine 5.

p3 = Machine 2 is idle and ready to process job 2. p4 = Job 2 is idle and ready to be processed by machine 2. p5 = Machine 3 is idle and ready to process job 3. p6 = Job 3 is idle and ready to be processed by machine 3. p7 = Job 1is ready to be processed by machine 4. p8 = Machine 4 is idle and ready to process job 1. p9 = Machine 4 is ready to process job 4. p10 = Job 4 is idle and ready to be processed by machine 4. p11 = Machine 5 is idle and ready to process job 4. p12 = Job 4 is ready to be processed by machine 5.

Table 4. Legends for the Petri Net graph (figure 3)

p1

p2

p10

p3

p4

p5

p6

t1 p7 p8 t4 p9

t2

t3

t5

p11 t6

p12

Figure 3. Petri Net graph for the

Ammunition Fuse shop. This Petri net (figure 3.0) shows the initial state and tokens in the form of dots represent the initial position of the resources. The Reachability matrix for the process is shown in table 5. It depicts the various states of the modeled net and the repetition of the first state in the end shows the net is deadlock free. P 1 2 3 4 5 6 7 8 9 10 11 12

S 1 2 .. N+1 1 1 1 1 1 1 0 1 0 1 1 0 1 1 1 1 1 1 1 0 1 1 .. .. .. .. .. .. .. .. .. .. .. 1 1 1 1 1 1 0 1 0 1 1 Table 5. Reachability Matrix for Petri net graph of Fuse shop 0 0 .. 0

ADVANTAGES AND LIMITATIONS OF PETRI NET Advantages The Petri Net modeling technique allows us to model systems which are concurrent, distributed, parallel, nondeterministic, and/or stochastic. The use of the Petri Net has the following advantages: 1. The transfer of resources or materials from one component of a system to another requires synchronised activities. This results in one component waiting for another. But in a Petri Net model , two events that are both enabled and do not interact may occur simaltaneously. So it is ideal for modeling systems of distributed control with multiple processes executing concurrently in time. 2. The pictorial description of a shop floor can be graphically represented in the Petri Net, which gives a clear idea about the sequence of operation required by each job. The Petri Net structure itself contains all the necessary information to define the possible sequence of events. Moreover the reachability matrix gives the states of each component of a

system during the process and can be used to find out the availability of the resouces at any instant. 3. Real time situations, like stalling of production can be checked in the Petri net by analysing the Reachability matrix and deadlock free processes can be designed. Rate of production Stochastic Petri Nets. 4. By using the Petri Net modeling, we can have an idea of status of production and it leads to control over the shop floor. If required other real life factors can also be included. Limitations The Classical Petri Net as developed by Carl Adam Petri carries some limitations with it. Research is underway to make Petri nets more real time oriented and to remove the restrictions in the modeling technique. Few of the limitations of the Petri Net are: 1. If the number of resources involved is large, then the Petri Net structure becomes very complex. To solve this complexity, some automation technique is required for making the Petri nets. 2. While developing classical Petri Nets it is considered that the processes are instantaneous and the events are primitive events. Since most events can also be found with the use of

in the real world take time, they are non primitive events and hence cannot be properly modeled by conventional method. 3. In a Petri net, the component route is fixed, so an alternative process path cant be followed which may exist in practice. This may lead to change in sequence of operation and further redesigning of Petri net and Reachability matrix.

6.0 FUTURE SCOPE OF PETRI NET The last decade has seen the application of Petri Nets do diverse fields and systems. A lot of research has been undertaken for the development of the classical model of Petri Net as defined by Carl. A. Petri. The inclusion of time in a Petri net has made the modeling more real life oriented and it can be used for quantitative analysis. Such Petri Nets are defined as Stochastic Petri Nets which includes the association of a firing delay with each transition. This delay specifies the time that the transition has to be enabled, before it can actually fire. When the delay is a random distribution function, the resulting net class is called stochastic Petri net. Moreover different types of transitions can be distinguished depending on their associated delay, for instance immediate transitions (no delay), exponential transitions (delay is an exponential distribution), and deterministic transitions (delay is fixed).

Another type of Petri Net which has developed more recently is the Colored Petri Nets. The concept of colored Petri Net was first introduced by Jensen in 1981. They are a high level Petri Net model that uses the concept of colors to increase its compactness. Research is still being undertaken to frame the rules for the application of colored Petri Nets. Efforts are being made to develop software which can generate the Petri Net model and Reachability matrix once the incidence matrix is fed to it. 7.0 CONCLUSION Manufacturing systems, where many processes evolve concurrently and share common resources, are indeed complex. Thus common modeling analysis and synthesis techniques are important issues for their design and operation. To keep pace with the changing world Petri Nets have come as an effective tool for performance evaluation and time control. The Petri net fully describes the concurrence of the processing operation. Its a graphical and a mathematical tool with the ability of visual communication. This graphic model shows the user the systems dynamic behavior and it can be modified to study various real time control concepts of the system. The application of Petri net can reduce the idle time for a process and leads to effective resource allocation and scheduling of jobs in a machine shop. In this case for the Fuse shop the analysis by Petri net has led to a much better jobmachine-job scheduling giving a more efficient machine utility and leading to drastic improvement in productivity.

REFERENCES:
1. Peterson J.L, Petri net theory and modeling of systems, pub. Prentice Hall

Inc. Englewood Clifts N.J. USA, 1981, pp. 1 106.

2. Narhari Y. and Vishwanathan N., Performance modeling of Automated

Manufacturing Systems, pub. Prentice Hall India Pvt. Ltd., 1994, pp. 487 546.
3. Proth J.M, Petri Nets: A Tool for Design and Management of Manufacturing

Systems, pub. John Wiley & Sons, Inc., 1997.

4. Singhee S and Pathak A.M, System modeling using Petri net, Project

report, KITS, Ramtek, 2002.

5.

Pachpor P.S, Petri net modeling of manufacturing systems with minimum duplicacy, 2000, pp. 11 25.