IAETSD JOURNAL FOR ADVANCED RESEARCH IN APPLIED SCIENCES ISSN NO: 2394-8442

A Simulation Modeling of Scheduling of Automated Guided Vehicles in

Flexible Manufacturing Systems Environment

Prof. Sumit Kumar Pandey1, Prof. Amit Sahay2, Prof. Virendra Rajput3
Mechanical Engineering Department, Mittal Institute Of Technology, Bhopal, Rajiv Gandhi Proudyogiki
Vishwavidhyalaya, Bhopal, Madhya Pradesh, India.

sumit263280@gmail.com [1]; amitsahaysolar@gmail.com [2]; virendra.rajput89@gmail.com [3]

ABSTRACT

Automated Guided Vehicles (AGVs) are among the fastest and advanced material handling technology that are used in
various industrial applications today. Flexible Manufacturing systems (FMS) are well suited for concurrent manufacturing
of a wide variety of parts in low quantity. The use of AGVs is increasing day by day for the material movement in
production lines of flexible manufacturing plants. The purpose is to increase efficiency in material transfer and increase
manufacturing. The order of scheduling of operations on machine centers as well as the scheduling of Automated Guided
Vehicles are important factors contributing to the overall efficiency of flexible manufacturing system (FMS). In this work,
scheduling of job is done for a particular type of flexible manufacturing system (FMS) environment by using the technique
of optimization called the genetic algorithm (GA). A code was developed to find the optimal solution and generate
random values in Ms-Excel. When a chromosome is input, the GA works upon it and produces same number of
offsprings. The number of iterations takes place until the optimum solution is obtained. Here we have worked upon eight
problems, with different no. of machines and no. of jobs. The input parameters used are Travel Time matrix and
Processing Time matrix with the number of machines and number of jobs. The results obtained are very quite close to the
results obtained by other techniques and by other scholars.

Keywords: AGVs; FMS; Make Span Time; DP; LP; GA; Chromosome.

1.0. INTRODUCTION
1.1. Present Status of AGVs:

The principal objectives of presents automation technology are flexibility and productivity, which can only be attained in
complete integrated manufacturing ambience. In this essential integration a conscientiously designed and sensibly arrange
material handling system is of importance. Automated guided vehicles (AGVs) are among the fastest increasing classes of
object in the material handling enterprise. They are computer-controlled driverless mobile vehicles having capabilities
for positioning and path selection. They are proficient of responding quickly to frequently changing transport model.
Automated guided vehicles (AGVs) can be merging too many other storage, production equipment and can be controlled
through an integrated computer control system. The requirement is to increase efficiency of production and material
transfer. However, while the hardware of AGVs has upgraded continuously, but the software of AGVs is still lack in many
respects. The area of operation of our AGVs is physically disjoint with the processed materials from these machining units
are output to separate AGV system.

VOLUME 4, ISSUE 6, NOV/2017 89 http://iaetsdjaras.org/

IAETSD JOURNAL FOR ADVANCED RESEARCH IN APPLIED SCIENCES ISSN NO: 2394-8442

In other words, we are only influenced about the distribution of raw materials from loading point (LPs) to various delivery
point (DPs) in a way that leads to best use of the machining units and the automated guided vehicles (AGVs). In material
transportation problem, the paths from Loading point (LPs) to Delivery point (DPs) are stick out like a tree with the
loading points (LPs) at the root and delivery point (DPs) on the branches. Because there are no closed loops, there are no
options about moving from the loading point (LPs) to any of the delivery point (DPs), or from one delivery point (DP) to
another delivery point. So the routing problem is very much solved in this case. We know the stock position of each
delivery Point (DPs) at any point of time. We have the data about average consumption rates of materials at each delivery
Point (DPs) from sensors mounted on the conveyors; we assume a particular load potential of the automated guided
vehicles (AGVs). In material transportation problem, the paths from Loading point (LPs) to delivery automated guided
vehicles (AGV).

1.2 Objective:

Our aim in examine an automated guided vehicle (AGVs) based flexible material distribution system which fits to
application in the following manner:

(1) Find out minimum how many AGVs will be necessary to meet the entire material distribution requirement.

(2) Suggest and judge miscellaneous dispatch rules for appointed transfer jobs to the automated guided vehicles
(AGVs). We input framework that prepare us to correlate acting of other dispatch rules in status of material
throughput and evenness of distribution over the delivery points (DPs).

(3) Then a pattern is reserved for partitioning out the complete area into special zones, one for each automated
guided vehicle (AGV) to trim the path among automated guided vehicles (AGVs) and thus avoid difficulties
arising out of that.

Here a try has been made to consider together the vehicle and machine scheduling aspects in a Flexible manufacturing
system (FMS) and address the issue for the minimization of make span time (Ms). Scheduling is influenced with the share
of limited resources to assignment over time and is a decision making process that correlates the time, operations, cost and
over all aim of the company.

2.0. METHODOLOGY
2.1 Operation and control flow:

A centralized processor monitors the material stock status of each DP conveyor. When the material stock at a DP falls
below a pre-specified level, it becomes eligible to receive materials. The central processor maintains a track of such eligible
DPs along with their current stock positions, so as soon as any AGV becomes available for service, it decides which DPs
should be served first, based on the prevailing dispatch rules. It can also determine the sequence in which the chosen DPs
may be visited to minimize the travel time. Accordingly it generates a transfer order for the AGV to execute. A transfer
order may contain a chain of nodes to be visited by the AGV. At some of these nodes, the AGV have to stop to load or
unload materials.

2.2 Role of Genetic Algorithm (GA):

The Genetic Algorithm (GA) is stochastic search procedure for combinatory optimization problems based on the
mechanism of natural selection and natural genetics. Essentially a Genetic Algorithm is a set of procedures that when
repeated, modify solutions for specific problem found.

VOLUME 4, ISSUE 6, NOV/2017 90 http://iaetsdjaras.org/

IAETSD JOURNAL FOR ADVANCED RESEARCH IN APPLIED SCIENCES ISSN NO: 2394-8442

Until a solution that yields acceptable results is found GAs generate successive populations of alternative solutions,
Individuals chromosomes are elected for reproduction, based on their fitness and the elected individuals, they undergo the
genetic operations, crossover and mutation. Few of the existing chromosomes of the current population are replaced with
the newly formed ones and this forms the population for the next generation and this process is continued till the
termination criterion is met. In this way a GA can quickly achieve a successful outcome without the need to examine every
possible solution to the problem. The time required for computation is very important, but the present-day speed of the
computer makes this process acceptable. The increasing performance of modern computer system permits the acceptable.
The increasing performance of modern computer systems permits the application of new algorithms such as GA and
simulated annealing, for production scheduling problems that requires too much computing time in the past.

3.0. PROBLEM FORMULATION AND SOLUTION

Typical operational, planning and control problems that have to be handled are:

Defining nodes and track-segments on the floor, with given no. of the LP, DPs, and accessible areas for vehicle
movement,

Assessing the number requirement of AGVs to meet the specified consumption rates at the DPs,

Defining parameters to evaluate and compare system performance under various dispatch rules,

Defining dispatch rules for the AGVs. A dispatch rule lays out a procedure, to decide which DPs may be served,
on the basis of current stock positions.

Routing of AGVs is done to ensure that they are utilized in an efficient manner. After the DP selection is over,
routing judges in which sequence the selected delivery points may be visited.

If there are a number of automated guided vehicle, they must obey some protocols for travelling over divided track
segments, to prevent getting into a deadlock situation. Here we subscribe to a decentralized control strategy, in our case, as
there are no alternative routes available to plan a trip on the basis of time windows.

3.1 Problem:

The FMS layout along with the distances between the machines and from the unload/load station are taken from Bilge
and Ulusoy (1995). The FMS consists of given no. of machines and 2 AGVs. The job set details and the distance between
two machines is given. AGV move with a speed of 40 m/min and the unloading and loading times of job are half min
each. The travel times are calculated and are presented in Tables, in which the unloading and loading times of the job are
included. Objective of problem: To Minimize the make span.

GA Parameters for problem:

Initial population size =4,

Probability of crossover=0.6,

Probability of mutation=0.01,

Number of generations=100 or a desirable value for objective, whichever occurs first?

VOLUME 4, ISSUE 6, NOV/2017 91 http://iaetsdjaras.org/

IAETSD JOURNAL FOR ADVANCED RESEARCH IN APPLIED SCIENCES ISSN NO: 2394-8442

Number of machines: 03
Number of jobs: 03

Table 1.1 Travel time matrixes:

L/U M1 M2 M3

L/U 0 2 4 10

M1 12 0 2 8

M2 10 12 0 6

M3 4 6 8 0

Table 1.2 Process time matrixes:

Job No. Job1 Job2 Job3

Machine M1 M3 M1 M2 M3 M2

Operation 1 2 3 4 5 6
Number

Table 1.3 Data for job set matrix:

Job M1 M2 M3

1 8 16 12

2 20 10 18

3 12 8 15

VOLUME 4, ISSUE 6, NOV/2017 92 http://iaetsdjaras.org/

IAETSD JOURNAL FOR ADVANCED RESEARCH IN APPLIED SCIENCES ISSN NO: 2394-8442

3.2 Calculation:

In GA genetic algorithm sequence of operation is generated which serves as the initial population. Let the initial
population of 4 sequences A, B, C, D is given in input.

Table 3.4 Initial population for problem

A 1 2 3 4 5 6

B 6 5 4 3 2 1

C 4 3 6 2 1 5

D 3 4 2 5 6 1

Calculation of make span time Ms for the chromosome A, B, C, D.

For Chromosome C: - 4 3 6 2 1 5

Operation 4:

Time transfer the job from L/U to machine 2 + Process time of job 2 on machine 2

4 + 10 = 14.

Operation 3:

Time transfer the job from machine 2 to machine 1 + Process time of job 2 on machine 1

Is 12 + 20 = 32.

Operation 6:

Time transfer the job from machine 1 to machine 2 + Process time of job 3 on machine 2

Is 2 + 8 = 10.

Operation 2:

Time transfer the job from machine 2 to machine 3 + Process time of job 1 on machine 3

Is 6 + 12 = 18.

Operation 1:

Time transfer the job from machine 3 to machine 1 + Process time of job 1 on machine 1

Is 6 + 8 = 14.

VOLUME 4, ISSUE 6, NOV/2017 93 http://iaetsdjaras.org/

IAETSD JOURNAL FOR ADVANCED RESEARCH IN APPLIED SCIENCES ISSN NO: 2394-8442

Operation 5:

Time transfer the job from machine 1 to machine 3 + Process time of job 3 on machine 3

Is 8 + 15 = 23.

Total make span time Ms for Chromosome C is = 14+32+10+18+14+23 = 111.

4.0. RESULT AND DISCUSSION

The result of the discussed above 2 problems with a nontraditional optimization approach using genetic algorithm (GA) is
applied to the given layouts and the set if input data. For the implementation of the algorithm the codes and there
randomization are developed in Ms- Excel. The criteria for stopping the running loop is either 100 generations or meeting
of the convergence criteria

Whichever occurs first? After meeting the earlier genetic algorithm (GA) parameters the codes are executed to give the
best routing sequences for the given FMS layout. The best chromosome results for the problem after certain number of
iterations is shown in Table 4.1 And graph 4.1 is plotted for showing the mean time (Ms) of different iterations of both the
problems which are the best chromosome for the Scheduling of AGVs in given FMS layout.

Table 4.1 Best Chromosomes for Problem

S.no Chromosome Make span (Ms)

1 A 109

2 B 117

3 C 111

4 D 103

Graph 4.1 Make Span Time values for Chromosomes

VOLUME 4, ISSUE 6, NOV/2017 94 http://iaetsdjaras.org/

IAETSD JOURNAL FOR ADVANCED RESEARCH IN APPLIED SCIENCES ISSN NO: 2394-8442

5.0. CONCLUSION
Optimization procedure has been developed in this work is based on genetic algorithm (GA) and is implemented and
successfully solved for the scheduling optimization problem of flexible manufacturing system (FMS). Codes are developed
and randomized in Ms-Excel. Results are obtained for the 2 problems having 3 machines and 4 machines respectively
flexible manufacturing system (FMS). With less calculated effort it is obtained the solution for the large number of
machines and jobs. This work leads to the conclusion that the procedure developed in this work can be suitably modified
to any kind of Flexible manufacturing system (FMS) with a large number of machines and components subjected to multi
objective functions.

REFRENCES
1. B.S.P. Reddy, C.S.P. Rao (2005), A Hybrid multi-objective GA for simultaneous scheduling of machines and AGVs in FMS,
Springer-Verlag London Limited 2006.

2. DeGuzman, M. C., Prabhu, N., & Tanchoco,

J. M. A. (1997), Complexity of the AGV shortest path and single loop guide path layout problems, International Journal of Production
Research, 35, 20832092.

3. Chandy, K., & Misra, J. (1979), Distributed simulation: A case study in design and verification of distributed programs, IEEE
Transactions on Software Engineering, 5, 440452.

4. Blazewicz J, Eiselt HA, Finke G, Laporte G, Weglarz J (1991), Scheduling tasks and vehicles in a flexible manufacturing system,
Int J Flex Manuf Syst 4:516.

5. Sabuncuoglu I, Hommertzheim DL (1992), Experimental investigation of FMS machine and AGV scheduling rules against the
mean flow time criterion. Int J Prod Res 30(7): 16171635.

VOLUME 4, ISSUE 6, NOV/2017 95 http://iaetsdjaras.org/