International Journal of Mechanical and Production

Engineering Research and Development (IJMPERD)

ISSN(P): 2249-6890; ISSN(E): 2249-8001
Vol. 8, Issue 6, Dec 2018, 465-478
© TJPRC Pvt. Ltd.

A GRAVITATIONAL SEARCH BASED ON THE OPTIMIZATION

OF JOB SHOP SCHEDULING PROBLEM

S. SRIDHAR1, A. JAWAHAR BABU2 & G. RANGAJANARDHANA3

1
Research Scholar, Department of Mechanical Engineering, JNTUK, Kakinada, Andhra Pradesh, India
2
Professor, Department of Mechanical Engineering, Gudlavalleru Engineering College, Andhra Pradesh, India
3
Professor, Department of Mechanical Engineering, JNTUA, Annanthapur, Andhra Pradesh, India
ABSTRACT

In a multi-factory optimization preparation, job allocation to a suitable factory or adjustable developed

component is a basic task. It encompasses the consideration of hardware implementation, technology, capability, and
an exploitation stage for each construction line or gathering component. The reduction of invention expenditure and
manufacturing capabilities are used to develop the competitiveness, an effectual construction system and also
diminishes the obtainable working time. The foremost target of this work is to diminish the overall order conclusion
time (i.e., makespan) by the aid of innovative Gravitational Search Algorithm related optimization procedure.
The anticipated method is executed on the functioning platform of MATLAB. The consequences were examined and

Original Article
distinguished by further obtainable methods.

KEYWORDS: Gravitational Search Algorithm (GSA); Genetic Algorithm (GA); Artificial Bee Colony (ABC); Job Shop
Scheduling Problem (JSSP); Job Shop Scheduling Problem (JSSP); Flexible Job-shop Scheduling Problem (FJSP)

Received: Aug 18, 2018; Accepted: Sep 12, 2018; Published: Nov 12, 2018; Paper Id.: IJMPERDDEC201850

1. INTRODUCTION

Generally, the liability and component alterations are contained by the real-world structure for to demand
job shop settings. According to the deviation in the shops, scheduling is difficult because scheduling decisions are
desiring to be prepared hurriedly [1]. The job-shop scheduling problem (JSP) is generally focused in the recent
tracks and it illustrates by the deliberation of scientists and specialists. The conventional JSP is generally
distinguished by no jobs. Here, each job contains a particular collection of functions. These functions are arranged
by means of the machines or work stations in a specified day and age which is pointed out by a specified particular
precedence position. A calendar is prepared to bind a computation (or numerous measures) of implementation [2].
Each job must undergo each machine at once. Each job is maintained during the machines in a precise demand and
there is no precedence necessities in the middle of diverse job function [3]. Each machine is must maintain a
solitary job promptly and it can't be interrupted. Moreover, the conducting time is established and identified.
The problem is to determine a schedule to bind the traverse.

Systematic solution approach is used to diminish the augmentation of difficulty size and the logically
anticipated shops are not expensive in a dynamic legitimate situation, in which, the modification of processes and
machines are modify the schedule [4]. In this manner, Operation Research (OR) is used to alternate the
transmitting principles or heuristics for to resolve the convenient occurrence in reasonable time. Additionally, the
problems of scheduling can be categorized as deterministic and nondeterministic classes which are derived from

www.tjprc.org SCOPUS Indexed Journal [email protected]

466 S. Sridhar, A. Jawahar Babu & G. Rangajanardhana

the quantity of vulnerability associated through the input limitations. The deterministic scheduling representation is not
including vulnerabilities rather than the actuality problems such as machine stoppage, diversity in due dates and
requirements, and irregular organizing times, which are disturbing the operational events [5]. Branch-and-bound (BB) or
Dynamic Programming (DP) is a prospective policy for discovery the accurate perfect preparations by means of supplying
diverse contender preparations in definite unreasonable condition. Though, these events are not capable and convincing in
managing the enormous size of problems. Suppose, if the amount of state features, is increased, then the DP faces the space
explosion problems [6].

Furthermore, the suitable transmitting regulations are selected by a heuristic and computerized analysis processes
which derives from shop situation. Suppose, if the regulation resolving methods are relying on contender regulations, then
it gives a hopeful conclusion which may not be available for a precise accumulating surroundings or a precise
implementation computation [7]. Some of the meta-heuristic procedures are GA, reproduced strengthening (SA), and
Cuckoo Search. In this process, an assembling structure was examining the problem of time-tabling. The analysed goal is
integrated the reduction of machine affecting time, stream time, and additionally construct negotiation. In this manner, the
problem of job-shop booking on solitary machines was inspected. Suppose, if these principles are available, then they are
not actually the finest ones [8]. An effectual island representation of GA (NIMGA) is exploited for JSSP and it exploits an
environment stimulated progression display and an additional truly provoked association resolve element, in which, each
island exploits an interchange conversion proprietor along with the self-adjustment phase of growth, and most remarkably
appalling populace through the least wellness progress initial to the adjacent islands observance in mind the target to
situate a better occasion to subsist in a more sensible island. NIMGA is a realistic representation of the environment than
the conventional IMGA, because it replicates the normal development in an enhanced manner [9, 10].

2. LITERATURE SURVEY

Hegen Xiong et al[11] have clarified a simulation-related examination of transmitting regulations for preparation
in an active job shop through batch discharge enchanting into a description of the comprehensive scientific superiority
limitation. This expression can be distinct as the expansion of predictable routing-related scientific superiority limitation.
Relating to tardiness values, the comparative presentations of few broadly employed transmitting regulations were
exploited. The efficiency of the four transmits regulations is exposing that the comparative presentation of transmitting
regulations could be influenced by means of several representation limitation. For the typical job shop preparation, the
difficulty representations have no comprehensive scientific preference, limitation among jobs, in addition to the
representation obtaining into describing the comprehensive scientific preference restriction. It was exposed to diminishing
the overall delay and the proportion of delayed jobs. The four transmitting regulations were very effectual. Concerning
delay-related intention, the comparative presentation of the examined transmitting regulations could be influenced by
means of altering the stage of the comprehensive scientific superiority limitation, but as well the due date stiffness.

Kameng Nip et al[12] have explained that the combinatorial advancement inconveniences were joining the great
open shop or occupation shop arrangement trouble and the most brief way trouble. It picks a division of employments that
contain a feasible clarification of the most limited way trouble. A while later, execute the favoured employments on the
shop machines to reduce the build traverse, i.e., the former accomplishment time of the whole occupations. They built up
that these inconveniences were NP-hard in two machines. Assume, if the amount of machines was an info, at that point
they represent that it was far-fetched to find harsh count calculations through introduction extent improved than 2 aside

Impact Factor (JCC): 7.6197 SCOPUS Indexed Journal NAAS Rating: 3.11
A Gravitational Search based on the Optimization 467
of Job Shop Scheduling Problem

from P = NP. They utilize an unconstrained figure calculation in the amount of machines as an info, and enhance guess
calculation in the amount of machines as foreordained. Moreover, they utilize a polynomial time guess organize for the
open shop condition in the amount of machine was foreordained.

Rui Zhang et al[13] have depicted an increasing apprehension in the ecological collision of conventional
manufacturing, particularly in expressions of energy utilization and the interrelated discharge of carbon dioxide. Moreover
the implementation of utensils, construction preparation could perform a significant function in dropping the overall energy
utilization of a developed plant. They clearly initiate the intention of diminishing energy utilization into a distinctive
construction preparation representation, i.e., the job shop scheduling difficulty is derived from the machine speed scaling
structure. To resolve that bi-intention optimization difficulty, they employ a multi-objective genetic algorithm integrated
through two difficulties-precise limited enhancement policies. These limited enhancement actions intend to augment the
explanation superiority through exploiting the arithmetical representation of two limited sub troubles founded on the
innovative difficulty.

Hao Gao et al[14] have anticipated a procedure for the activity shop booking issue (JSSP) which got from the
combination meta heuristic process. This procedure is utilized for the characteristics of an improved molecule swarm
streamlining (PSO) and a Tabu pursuit (TS) calculation. It was utilized to inspect an essential segment methodically.
The rest of the approach was started in the PSO for building up its examination capacity. Subsequently, the improved PSO
could supply different and best preparatory clarification to the TS for creation a prevalent explores in the overall space.
They likewise use a constrained look approach for gaining improved outcome in JSSP. A genuine whole number encodes
and interpret, organize for relating a clarification in consistent space to a segregated arrangement clarification was planned
for the upgraded PSO and the Tabu calculation to transparently relate their clarification for becoming the research of
improved clarification.

S. Meeran et al[15] have clarified a Job Shop Scheduling trouble by methods for a various assortment of utensils
and systems, for example, Branch and Bound at one final piece of the field and Heuristics at the supplementary end.
However, the exposition evaluations are suggested that these techniques were satisfactory on their person to determine that
adamant NP-hard trouble. In this manner, it was propose that a legitimate clarification process needed to build up the
noteworthy characteristic of various arrangements. Right now they gave one such clarification procedure to coordinating
Genetic Algorithm and Tabu Search. The establishment of such combination process as in the state of further a framework
which utilize GA and TS was to join the extension all inclusive inquiry and develop constrained hunt fitness of GA and TS
correspondingly. The fusion representation employed as exceed in many comparable systems of conventional standard
troubles and real-life troubles. In this process, the system accomplish the mutual collision of numerous diminutive but
significant attribute such as dominant chromosome depiction, effectual genetic operatives, limited neighborhood policies
and competent search policies together with original preliminary explanation. The system’s sensible convention features
are exhibit that the system was regularly competent of resolving the actual life Job Shop troubles.

Yoni. Nazarathy et al[16] have declared an extraordinary amount of employment shop booking inconveniences, in
which a foreordained amount of machines, a delimited amount of activities per work, and an awesome amount of
occupations were actualized. In the gigantic amount of occupation shops, it builds to determine a liquid trouble and to
program the employments in such a way as to take after the liquid clarification. There had been various records which
utilized assessed clarifications were asymptotically most positive as the amount increases. It was unspecified that the

www.tjprc.org SCOPUS Indexed Journal [email protected]

468 S. Sridhar, A. Jawahar Babu & G. Rangajanardhana

difficulty contains several indistinguishable copies of a predetermined group of jobs. They employ a very undemanding
heuristic which could program such troubles. They converse asymptotic optimality of this heuristic, beneath a broad
collection of formerly uncultivated circumstances.

3. PROPOSED GSA BASED JSSP METHOD

The Flexible Job shop Scheduling Problem (FJSP) is appropriate for the conventional job scheduling difficulty
which is overcome by whichever machine contains pre-arranged preparation. The correctly systematized functions of the
machines are helping to diminish the necessity of working time in the JSSP procedure, in order to accomplish the
consumer requirements The fundamental capability is to distribute each solitary task to a machine and to compile the
capability on the machines, every single one together that the maximum expression of consummation (make span) of the
entire task is reduced to the bottom. Each procedure is to be switched by means of diverse machines, in a specified
demand. The problem is to determine a program of the irrelevant time to terminate the entire jobs. Therefore, the
distinguishing standard troubles are measured to diminish the construct duration time of the complete mission by means of
Gravitational search Algorithm (GSA) related optimization procedure. The anticipated Optimization progression reflects
on the intention task (i.e., fitness function) because to reduce the make span time. Suppose, if the preferred make span time
or greatest iteration is accomplished, then the system is premeditated to conclude.

3.1 System Model Formulations and Assumptions

Here, if the activity obtains diverse jobs, then it requests to allocate the jobs to a logical factor in the intervening
time. The scheme of job preparation must terminate the complete task, at least span time. The fundamental intention of this
process is to augment the implementation efficiency. Jobs are premeditated by several optimization principles and the
produced procedures are evaluated for various implementation conditions like the smallest amount of expenditure, slightest
make span time, slightest tardiness, severe resource exploitation, and so on.

In this paper, the ( j × m ) is indicating the least make span common job shop scheduling difficulty, that can be

signified by means of a group of ‘ j ’ jobs, S J . The managing of j th job ( S Jj ) on the m th machine ( S Fm, M ) is

recognized as the function, S Jo ,O . Function S Jo ,O necessitates the discriminating exploitation of machine S Fm, M for a

constant period t j ,o , m (i.e. processing time). A plan is a group of overriding times for each function that gratify the

intention task. The purpose of the anticipated optimization difficulty (i.e. objective function) is to bound the make span
time. The make span time T is the mandatory time to terminate every single one of the jobs. The preparation difficulty is
compound, because it requires adjusting the particular complexity of ( j × m) least makespan common job shop

preparation difficulty. The common job shop difficulty contains ‘ j ’ jobs and ‘ m ’ machines, therefore the quantity of

reasonable explanation is ( j!) .

m

3.1.1 Formulations (Input Parameters)

Let the set of factories, machines, jobs and its operations be represented by the following equations (1), (2), (3),
(4) respectively.

{
S F = S F1 , S F2 , … , S Ff } (1)

Impact Factor (JCC): 7.6197 SCOPUS Indexed Journal NAAS Rating: 3.11
A Gravitational Search based on the Optimization 469
of Job Shop Scheduling Problem

{
S F , M = S F1 , M , S F2 , M , … , S Fm , M } (2)

{
S J = S 1J , S J2 , … , S Jj } (3)

{
S J , O = S 1J , O , S J2 , O , … , S Jo , O } (4)

Where,

f th
S F - set of factories and S F denotes the f factory

S J - set of jobs and S Jj denotes the j th job

th
S J ,O - number of operations in j job ‘ S Jj ’ and S Jo ,O denotes the o th operation

S F , M - set of machines and S Fm, M denotes the m th machine

Moreover, some other parameters related to time that are essential in computing the objective function are given
as follows,

T j , f -Deadline time for delivering a job ‘ j ’ from factory ‘ f ’ to the hub

T jD,o , f ,m - Total time duration required for any operation ‘ o ’ of a job ‘ j ’ in a machine ‘ m ’ from the set of

factories S F

T jS, o , f , m - starting time of an operation ‘ o ’ of a job ‘ j ’ in a machine ‘ m ’ from the set of factories S F

T jB, o , m - Beginning time of any operation ‘ o ’ of a job ‘ j ’ in the machine ‘ m ’

T jE, o , m - Ending time of any operation ‘ o ’ of a job ‘ j ’ in the machine ‘ m ’

The step by step procedure taking place in the GSA procedure is given as,

3.1.2 Assumptions

j
The job S J encompasses SO group of functions, which must be maintained by means of a predefined

progression.

• At some point in the function of a machine, apart from of collapse, injure, or substance insufficiency, once each
function is in line, neither discontinue movement nor reprocessing is acceptable.

• The implementation of function S J , O ∈ S O , S O = S 1O , S { 2

O }
, … , S Oo of job, S J | S J = {S J1 , S J2 , … , S Jj }

preferred from a group of obtainable machines, S F , M .

• Each one machine can carry out a solitary function S j , o at some specified time and the transportation times

among machines are ignored.

• Each function of a job must be prearranged to one solitary factory to allocate the job.

www.tjprc.org SCOPUS Indexed Journal [email protected]

470 S. Sridhar, A. Jawahar Babu & G. Rangajanardhana

3.1.3 Objective Function (Fitness Function)

The objective function is the minimum of makespan time, T , that can be formulated as,

 
(
Fitness = Min (T ) = Min Max ∑ ∑ T jS,o , f ,m + T jD,o , f ,m + T j , f  )
 

For reaching the objective function, the proposed JSSP introduces Gravitational Search Algorithm based
optimization procedure.

3.2 Gravitational Search Optimization Algorithm

The Gravitational Search Algorithm (GSA) is derived from the activities of Newton’s gravitational force which
was anticipated by means of Rashedi. In this algorithm, mediators are indicated as models and their presentation is
considered by means of their loads. Each model is magnetizing supplementary by the help of gravitational force, and this
force origin a comprehensive progress of the entire model in the direction of the model through heavier masses. Hereafter,
loads exploiting a straight structure of communication by the aid of gravitational force. At GSA, each one mass (agent)
encompasses four terms such as the location, the inertial mass, active gravitational mass and the passive gravitational mass.
For each explanation, the gravitational and inertial masses are premeditated by means of fitness task. Based on the loads,
force and accelerations are calculated. At last, the velocity and location of the mediators are modernized by the stepping up
elements. The entire limitations are modernized in termination.

Here, the flow diagram of the GSA optimization procedure is depicted in the following figure 1.

Step 1: Initialization

Based on the quantity of functions and the reasonable machines, input populace ‘ M ’ is erratically established by

means of unsystematic location athe mediatorois signified j th mediator is signify by the subsequent association:

Yj = (y 1
j , y 2j , . . . , y nj , . . . , y mj ) for = ( j = 1, 2 , . . . , M ) (5)

Where, y nj is the position of j th agent value in the n dimension

Impact Factor (JCC): 7.6197 SCOPUS Indexed Journal NAAS Rating: 3.11
A Gravitational Search based on the Optimization 471
of Job Shop Scheduling Problem

Figure 1: Flow Diagram for GSA

Step 2: Fitness Calculation

In this step, the fitness evaluation is done by utilizing the objective function.

Fitness p (k ) = Min (T ) (6)

Where, Fitness (k ) indicates the fitness of p th mediator at k th iteration. The fitness value is declared as

Fitness (k ) , the finest and nastiest fitness for the entire mediator are estimated at quantity of iteration. The finest and

www.tjprc.org SCOPUS Indexed Journal [email protected]

472 S. Sridhar, A. Jawahar Babu & G. Rangajanardhana

nastiest fitness are declared as:

Fitnessbest (k ) = max Fitness p (k ) (7)

p∈(1,..,M )

Fitnessworst (k ) = min Fitness p (k ) (8)

p∈(1,..,M )

Step 3: Gravitational Constant ( C ) Computation

The gravitational constant C is located at the establishment which is measured by a phase of time to accomplish
the search accuracy. The gravitational constant is specified as,

 k 
 −λ 
C (k ) = C0 e  K
(9)

Where, K is the maximum iteration and λ and C0 are constants

Step 4: Update C , Best and Worst Fitness

For the period of iteration, the finest and nastiest fitness values were modernized occasionally in the Gravitation
constant which is derived from the preceding explanation.

Step 5: Gravitational and Inertial Masses Computation

According to the fitness values, the Gravitational and inertia masses were premeditated. A heavier load illustrates
an enormously competent explanation. Presuming the correspondence of the gravitational and inertia mass, the values of
load is evaluated by means of the plan of fitness.

G aj = G pj = G ij = G j , where (j = 1, 2 , . . . , M )

The mass of j th agent is calculated as,

g j (k )
G j (k ) = M

∑ g (k )
p =1
p

In the above equation, g j (k ) can be obtained as,

Fitness (t )− worst (t )
g j (k ) = j
(10)
best (t ) − worst (t )

Where, G aj , G pj , Gij and G j are the active gravitational mass, passive gravitational mass, inertial gravitational

mass and the massof j th agent. Also, Fitness j (t ) signifies the fitness value of j th agent at k th iteration.

Step 6: Total Force and Accelerations Calculation

Then the force and acceleration values of the j th agent is determined, that can be expressed as:

Impact Factor (JCC): 7.6197 SCOPUS Indexed Journal NAAS Rating: 3.11
A Gravitational Search based on the Optimization 473
of Job Shop Scheduling Problem

Force nj (k ) = ∑ rand p Force njp (k )

p∈qbest , p ≠ j

Force nj (k )
Acc nj (k ) = (11)
Gij (k )

Where, Force n
jp (k ) is the force acting on the j th agent by p th agent.

Step 7: Update Velocity and Position

Moreover, the established velocity is anticipated by means of the regulation of movement. The velocity
calculation is denoted by the ensuing equation:

V jn ( k + 1) = rand j × V jn ( k ) + Acc nj (k ) (12)

Where, rand i denotes the uniform random variable in the interval [0, 1]

At present, the location of the mediator will be restructured by means of the anticipated velocity and the location
renewing is organized by means of the ensuing equation:

Position nj ( k + 1) = Position nj ( k ) + V jn ( k + 1) (13)

Step 8: Repeat

Reiterate the procedure up to the best solution is attained.

Step 9: Terminate

Subsequent to accomplishing the finest explanation through negligible makespan time, the quantity of iteration is
concluded.

4. RESULTS AND DISCUSSIONS

This segment explains about the result and discussion of an effectual Job Shop Scheduling difficulty by means of
GSA related optimization procedure. The anticipated, algorithm is accomplished by the aid of MATLAB software and the
testing is taking place by a system of encompassing 4 GB RAM and 2.10 GHz Intel I-3 processor.

The effectiveness of the anticipated GSA related optimization algorithm on JSSP is investigated by means of 20
standard troubles of unreliable size. This obtained data is exploited in estimating the Makespan time of diverse
optimization related JSSP. The investigational consequences of diverse optimization related JSSP were examined in this
segment. Additionally, evaluation is prepared for the typical Makespan time and the replicated consequences.

Table 1 gives the original and simulated Makespan Time values (in seconds) for a number of Benchmark
problems.

www.tjprc.org SCOPUS Indexed Journal [email protected]

474 S. Sridhar, A. Jawahar Babu & G. Rangajanardhana

Table 1: Make Span Time for a Number of Benchmark Problems

Benchmark Size Make Span Time (Seconds)

Problems ( j × m) Original GSA ABC GA
FT10 6 × 12 930 887 909 928
LA01 10 × 20 666 635 641 642
LA02 10 × 10 655 602 641 643
LA03 10 × 10 597 563 573 592
LA04 10 × 10 590 558 569 571
LA05 10 × 10 593 560 582 585
LA06 15 × 10 926 843 869 881
LA07 15 × 10 890 844 877 882
LA08 15 × 10 863 812 842 862
LA09 15 × 10 951 916 918 951
LA10 15 × 10 958 910 931 950
LA11 20 × 10 1222 1163 1167 1200
LA12 20 × 10 1039 981 998 1028
LA13 20 × 10 1150 1106 1110 1146
LA14 20 × 10 1292 1207 1220 1245
LA15 20 × 10 1207 1148 1168 1200
LA30 20 × 20 1355 1341 1343 1355
LA31 30 × 20 1784 1758 1769 1777
LA33 30 × 20 1719 1691 1719 1719
LA35 30 × 20 1888 1866 1866 1868

From the above table,

• It is distinguished that the least Makespan time (i.e. 558sec) is accomplishing for the standard difficulty ‘LA04’ of

size ( 10 × 10 ) for the anticipated GSA optimization related JSSP. Additionally, it is better than 568 se for the
entire supplementary process.

• Moreover, it can be observed that the Makespan time is condensed for the anticipated method even if the size of
the standard difficulty is elevated.

• Since the acquired consequences, the replicated consequences in the course of optimization related JSSP are
enhanced than the innovative Makespan time values.

• Additionally, the evaluation of optimization related JSSP methods, the GSA related JSSP provides the smallest
amount of Makespan time.

• The ABC related JSSP as well offers considerable consequences, but the replicated consequences for definite
standard troubles were not indicating the spot. Therefore the anticipated GSA related JSSP locate frontward than
the entire further methods.

Additionally, the evaluation scheme for the unique in addition to the replicated Makespan time values are
calculated with the help of anticipated GSA optimization related JSSP and further optimization related JSSP such as ABC,
GA is specified in the beneath figures 2 to 5.

The figure 2 is used to illustrate the innovative in addition to the replicated Makespan time values for the standard
troubles like FT10, LA01, LA02, LA03 and LA04.

Impact Factor (JCC): 7.6197 SCOPUS Indexed Journal NAAS Rating: 3.11
A Gravitational Search based on the Optimization 475
of Job Shop Scheduling Problem

Figure 2: Make span time for 5 benchmark problems (FT10, LA01, LA02, LA03, LA04)

For the standard troubles like LA05, LA06, LA07, LA08 and LA09, the innovative and the replicated Makespan
time values were conspired in figure 3.

Figure 3: Make Span Time for 5 Benchmark Problems (LA05, LA06, LA07, LA08, LA09)

The figure 4 and 5 explains about the innovative and the replicated Makespan time values for the group of
standard troubles (LA10, LA11, LA12, LA13, LA14) and (LA15, LA30, LA31, LA33, LA35) correspondingly.

Figure 4: Make Span Time for 5 Benchmark Problems (LA10, LA11, LA12, LA13, LA14)

www.tjprc.org SCOPUS Indexed Journal [email protected]

476 S. Sridhar, A. Jawahar Babu & G. Rangajanardhana

Figure 5: Make Span Time for 5 Benchmark Problems (LA15, LA30, LA31, LA33, LA35)

5. CONCLUSIONS

This document is used to resolve the job shop scheduling difficulty with the help of GSA related optimization
procedure. At this point, the accomplishment is prepared with the help of twenty rough standard occurrences and the
replication consequences were effectively achieved and examined for further obtainable optimization related JSSP methods
predominantly, ABC and GA related optimization process. The entire investigation consequences are illustrated that the
anticipated GSA optimization related JSSP surpass supplementary process for the entire standard troubles.

REFERENCES

1. A. Azadeh , N. Hosseini , S. Abdolhossein Zadeh and F. Jalalvand, "A hybrid computer simulation-adaptive neuro-fuzzy
inference system algorithm for optimization of dispatching rule selection in job shop scheduling problems under uncertainty",
The International Journal of Advanced Manufacturing Technology, vol. 79, no. 1-4, pp. 135-145, 2015.

2. Mellado, C. Cubillos and D. Cabrera, "A constructive heuristic for solving the Job-Shop Scheduling Problem", IEEE Latin
America Transactions, vol. 14, no. 6, pp. 2758-2763, 2016.

3. H. S. Keesari and R. V. Rao, "Optimization of job shop scheduling problems using teaching-learning-based optimization
algorithm", OPSEARCH, vol. 51, no. 4, pp. 545-561,2013.

4. Lvjiang Yin, Xinyu Li, Liang Gao, Chao Lu and Zhao Zhang, "A novel mathematical model and multi-objective method
for the low-carbon flexible job shop scheduling problem", Sustainable Computing: Informatics and Systems, 2016.

5. Sadegh Mirshekarian and Dusan N.Sormaz, "Correlation of job-shop scheduling problem features with scheduling efficiency",
Expert Systems with Applications, vol. 62, pp. 131-147, 2016.

6. Liping Zhang, Liang Gao and Xinyu Li, "A hybrid intelligent algorithm and rescheduling technique for job shop scheduling
problems with disruptions", The International Journal of Advanced Manufacturing Technology, vol. 65, no. 5-8, pp. 1141-
1156, 2012.

7. Mohamed Kurdi, "An effective new island model genetic algorithm for job shop scheduling problem", Computers &
Operations Research, vol. 67, pp. 132-142, 2016.

8. Leila Asadzadeh, "A parallel artificial bee colony algorithm for the job shop scheduling problem with a dynamic migration
strategy", Computers & Industrial Engineering, vol. 102, pp. 359-367, 2016.

Impact Factor (JCC): 7.6197 SCOPUS Indexed Journal NAAS Rating: 3.11
A Gravitational Search based on the Optimization 477
of Job Shop Scheduling Problem

9. Su Nguyen,Mengjie Zhang,Mark Johnston and Kay Chen Tan, "Automatic Programming via Iterated Local Search for
Dynamic Job Shop Scheduling", IEEE Transactions on Cybernetics, vol. 45, no. 1, pp. 1-14, 2015.

10. Omid Gholami and Yuri N. Sotskov, "A fast heuristic algorithm for solving parallel-machine job-shop scheduling problems",
The International Journal of Advanced Manufacturing Technology, vol. 70, no. 1-4, pp. 531-546, 2013.

11. Hegen Xiong , Huali Fan , Guozhang Jiang and Gongfa Li, "A simulation-based study of dispatching rules in a dynamic job
shop scheduling problem with batch release and extended technical precedence constraints", European Journal of
Operational Research, vol. 257, no. 1, pp. 13-24, 2017.

12. Saravanan, N., & Rathnasamy, R. A New Tool using Simulation and Optimization of Solar Adsorption Cooling Technique.

13. Kameng Nip, Zhenbo Wang and Wenxun Xing, "A study on several combination problems of classic shop scheduling and
shortest path", Theoretical Computer Science, vol. 654, pp. 175-187,2016.

14. Rui Zhang and Raymond Chiong, "Solving the energy-efficient job shop scheduling problem: a multi-objective genetic
algorithm with enhanced local search for minimizing the total weighted tardiness and total energy consumption", Journal of
Cleaner Production, vol. 112, pp. 3361-3375, 2016.

15. Hao Gao, Sam Kwong, and Ran Wang, "A Hybrid Particle-Swarm Tabu Search Algorithm for Solving Job Shop Scheduling
Problems", IEEE Transactions on Industrial Informatics, vol. 10, no. 4, pp. 2044-2054, 2014.

16. S. Meeran and M. S. Morshed, "A hybrid genetic tabu search algorithm for solving job shop scheduling problems: a case
study", Journal of Intelligent Manufacturing, vol. 23, no. 4, pp. 1063-1078, 2011.

17. Yoni. Nazarathy and Gideon. Weiss, "A fluid approach to large volume job shop scheduling", Journal of Scheduling, vol. 13,
no. 5, pp. 509-529, 2010.

www.tjprc.org SCOPUS Indexed Journal [email protected]