a multi-objective bisexual reproduction genetic algorithm for ...

More documents

Recommendations

Info

51 The new offspring is created from an empty chromosome, and then it is inserted alternately with genes from mother and father. Because a n-credit course will be scheduled to n successive time-slots, successive time-slots have to be copied from mother and father. To facilitate this, all time-slots of morning or afternoon working sessions will be copied from the mother or father to the new offspring. Usually the new offspring is not correct thus it needs to be repaired. If a course has not been scheduled yet, it also needs to be scheduled. In the contrary, if a course has been scheduled more than one time in a week, it has to be removed. A pseudo code for crossover is given in Figure 3-20. Crossover rate pc=0.5 Father x= a chromosome is chosen randomly from the population Mother y= a chromosome is chosen randomly from the population (y≠x) For each day in a week For each working-session in [morning, afternoon] For each classroom If (random(100) < pc*100) Copy afternoon time-slots of father x to afternoon time-slots of the new offspring z Else Copy morning time-slots of mother y to morning time-slots of the new offspring z Mutate the new offspring z Repair the new offspring z Calculate fitness value for the new offspring z Insert the new offspring z into the population in order of fitness value FIGURE 3-20 Pseudo code for crossover If the crossover rate pc is chosen to be 50%, the 50% of the genes from the mother and 50% of the genes from father are copied to the new offspring. 3.6.6 Mutation A new offspring that has just been created by crossover will be mutated with a mutation rate. This is done via the following process: go through each gene and swap its content with another gene in the same chromosome.
52 As mentioned in the previous section, a course has to be scheduled to successive time-slots, so we have to swap the successive time-slots booked for a course with other successive time-slots. To facilitate this, we choose all time-slots of a working session to swap with those of another, as illustrated in Figure 3-21. … … Classroom j Mon Tue Wed Thu Fri Chromosome x Course 6 Course 8 Course 6 Course Classroom i 8 Course6 Course 8 Mon Tue Wed Thu Fri Course 1 Course 2 Course 1 Course 2 Course 1 Course 2 Course3 Course 9 Course3 Course 9 Course 3 Course 3 Course 4 Course 4 Swap contenst of 2 workingsessions with each other FIGURE 3-21 Mutation A pseudo code for mutating is given in Figure 3-22. Mutation rate pm=0.02 For each classroom For each day in a week For each working-session in [morning, afternoon] If (random(100) < pm*100) R= a classroom is chosen randomly from the classroom group that is the same group of the current classroom Swap all time-slots of the current working-session of the current classroom with those of classroom R FIGURE 3-22 Pseudo code for mutating a chromosome Because a course is scheduled by only using classrooms in an assigned classroom group, any swapping has to ensure to use the classrooms within this classroom group.
Page 1 and 2:
COURSE SCHEDULING IN MULTIPLE FACUL
Page 3 and 4:
ชื่อ : นายฮูเ
Page 5 and 6:
TABLE OF CONTENTS Page Abstract (in
Page 7 and 8: LIST OF TABLES Table Page 2-1 Cours
Page 9 and 10: LIST OF FIGURES (CONTINUED) Figure
Page 11 and 12: 2 The classrooms are sometime share
Page 13 and 14: 4 Classroom i Time-slot Hour Mon Tu
Page 15 and 16: CHAPTER 2 LITERATURE REVIEW In this
Page 17 and 18: 9 teach effectively. However, in so
Page 19 and 20: 11 The set of vertexes are consider
Page 21 and 22: 13 Gueret et al. have implemented a
Page 23 and 24: 15 When the atoms are at a high tem
Page 25 and 26: 17 The outline of the basic genetic
Page 27 and 28: 19 As shown in Figure 2-6, first of
Page 29 and 30: 21 2.3.2.1 Crossover After we have
Page 31 and 32: 23 these chromosomes. According to
Page 33 and 34: 25 2.4.1 Application Considerations
Page 35 and 36: 27 Southern California, the Argonne
Page 37 and 38: 29 2.4.3.2 Application Programming
Page 39 and 40: 31 2.5 Summary The course schedulin
Page 41 and 42: 34 3.1.5 Phase 5: Measurement a) To
Page 43 and 44: 36 All the above problems can be pr
Page 45 and 46: 38 Client at a faculty Client at th
Page 47 and 48: 40 The Grid system is only used by
Page 49 and 50: 42 3.6 The Proposed Genetic Algorit
Page 51 and 52: 44 Chromosome x i Fitness = f(x i )
Page 53 and 54: 46 A population is a list of n chro
Page 55 and 56: 48 Lecturers register their busy ti
Page 57: 50 A course is booked to the time-l
Page 61 and 62: 54 GIIS from the client machine m1.
Page 63 and 64: 56 The complete source code for man
Page 65 and 66: 58 3.8.2.2 Stage 2 After the centra
Page 67 and 68: 60 3.8.3.3 The Gatekeeper The gatek
Page 69 and 70: 62 TABLE 4-1 (CONTINUED) Class Seme
Page 71 and 72: 64 26 lecturers are assigned to tea
Page 73 and 74: 66 TABLE 4-2 (CONTINUED) Course Sec
Page 75 and 76: 68 This result shows that the GA ra
Page 77 and 78: 70 should revise the chromosome rep
Page 79 and 80: 72 - Population size : 10 - Crossov
Page 81 and 82: 74 for network communication is muc
Page 83 and 84: 76 TABLE 4-5 Timetable created by t
Page 85 and 86: CHAPTER 5 CONCLUSION 5.1 Conclusion
Page 87 and 88: REFERENCES 1. Alkan, A. and Ozcan,
Page 89 and 90: 83 20. Burke, E. K. and Newall, J.
Page 91 and 92: 85 38. Globus Toolkit [Online]. Ava
Page 93 and 94: 88 This section presents the struct
Page 95 and 96: 90 A.5 Building TABLE A-5 Building
Page 97 and 98: 92 A.10 Program TABLE A-10 Program
Page 99 and 100: 94 A.14 Timetable TABLE A-14 Timeta
Page 101 and 102: 96 This section presents in detail
Page 103 and 104: 98 TABLE B-2 Group, user ID and pas
Page 105 and 106: 100 broadcastdelay 0.008 authentica
Page 107 and 108: 102 mycapw [enter] During the above
Page 109 and 110:
104 service gsigatekeeper { socket_
Page 111 and 112:
106 B.3.3 Starting the MDS on All o
Page 113 and 114:
108 B.4 Checking the Installation T
Page 115 and 116:
110 This section introduces the ste
Page 117 and 118:
112 C.3.2 Configuration This sectio
Page 119 and 120:
114 [root@m1 root]#grid-proxy-info
Page 121 and 122:
116 The system will automatically p
Page 123 and 124:
118 C.6 Installing JDBC Driver on L
Page 125 and 126:
120 This section presents how to co
Page 127 and 128:
122 All the following files are com
Page 129 and 130:
124 } } } try { ctx.close(); } catc
Page 131 and 132:
126 }//for }else if(attribute.equal
Page 133 and 134:
128 public static void main(String
Page 135 and 136:
130 System.out.println(CentralizedS
Page 137 and 138:
132 gassJob[jobCount]=new GassJob(t
Page 139 and 140:
134 /** * Init job out listeners fo
Page 141 and 142:
136 // Start GASS server if (! star
Page 143 and 144:
138 // waits for DONE or FAILED sta
Page 145 and 146:
140 //get RSL of Job having index i
Page 147 and 148:
142 class Job { private String jobn
Page 149:
145 BIOGRAPHY Name : Mr. Nguyen Con
show all

a multi-objective bisexual reproduction genetic algorithm for ...

Create successful ePaper yourself

Delete template?

Save as template?