scholarly journals Monthly Rainfall Prediction Model of Peninsular Malaysia Using Clonal Selection Algorithm

2018 ◽  
Vol 7 (4.35) ◽  
pp. 182
Author(s):  
N.S.Noor Rodi ◽  
M.A. Malek ◽  
A.R. Ismail
Keyword(s):  
Immune System ◽  
Clonal Selection ◽  
Peninsular Malaysia ◽  
Monthly Rainfall ◽  
Rainfall Data ◽  
Series Data ◽  
Monthly Interval ◽  
Clonal Selection Algorithm ◽  
Selection Algorithm ◽  
Rainfall Prediction

Nowadays, various algorithms inspired by natural processes have been extensively applied in solving engineering problems. This study proposed Artificial Immune Systems (AIS), a computational approach inspired by the processes of human immune system, as an algorithm to predict future rainfall. This proposed algorithm is another alternative technique as compared to the commonly used Statistical, Stochastic and Artificial Neural Network techniques traditionally use in Hydrology. Rainfall prediction is pertinent in order to solve many hydrological problems. The proposed Clonal Selection Algorithm (CSA) is one of the main algorithms in AIS, which inspired on Clonal selection theory in the immune system of human body that includes selection, hyper mutation, and receptor editing processes. This study proposed algorithm is utilised to predict future monthly rainfall in Peninsular Malaysia. The collected data includes rainfall and other four (4) meteorological parameters from year 1988 to 2017 at four selected meteorological stations. The parameters used in this analysis are humidity, wind speed, temperature and pressure at monthly interval.  Four (4) meteorological stations involved are Chuping (north), Subang Jaya(west), Senai (south) and Kota Bharu (west) represented peninsular Malaysia. Based on the results at testing stage, it is found that the trend and peaks of the hydrographs from generated data are approximately similar to the actual historical data. The highest similarity percentage obtained is 91%. The high values of similarity percentage obtained between simulated and actual rainfall data in this study, reinforced the hypothesis that CSA is suitable to be used for prediction of continuous time series data such as monthly rainfall data which highly variable in nature.  As a conclusion, the results showed that the proposed Clonal Selection Algorithm is acceptable and stable at all stations.

A clonal selection algorithm model for daily rainfall data prediction

2014 ◽  
Vol 70 (10) ◽  
pp. 1641-1647
Author(s):  
N. S. Noor Rodi ◽  
M. A. Malek ◽  
Amelia Ritahani Ismail ◽  
Sie Chun Ting ◽  
Chao-Wei Tang
Keyword(s):  
Clonal Selection ◽  
Daily Rainfall ◽  
Rainfall Data ◽  
Series Data ◽  
Clonal Selection Algorithm ◽  
Selection Algorithm ◽  
Immune Systems ◽  
Daily Rainfall Data ◽  
Data Prediction ◽  
Testing Stage

This study applies the clonal selection algorithm (CSA) in an artificial immune system (AIS) as an alternative method to predicting future rainfall data. The stochastic and the artificial neural network techniques are commonly used in hydrology. However, in this study a novel technique for forecasting rainfall was established. Results from this study have proven that the theory of biological immune systems could be technically applied to time series data. Biological immune systems are nonlinear and chaotic in nature similar to the daily rainfall data. This study discovered that the proposed CSA was able to predict the daily rainfall data with an accuracy of 90% during the model training stage. In the testing stage, the results showed that an accuracy between the actual and the generated data was within the range of 75 to 92%. Thus, the CSA approach shows a new method in rainfall data prediction.

scholarly journals Artificial Immune System Applied to Job Shop Scheduling

2021 ◽  
pp. 15-22
Author(s):  
Steven Kosasih ◽  
◽  
Cecilia E. Nugraheni ◽  
Luciana Abednego
Keyword(s):  
Immune System ◽  
Lower Bound ◽  
Upper Bound ◽  
Job Shop ◽  
Job Shop Scheduling ◽  
Clonal Selection ◽  
Clonal Selection Algorithm ◽  
Selection Algorithm ◽  
Artificial Immune

Job Shop Scheduling is a problem to schedule n number of jobs in m number of machines with a different order of processing. Each machine processes exactly one job at a time. Each job will be processed in every machine once. When a machine is processing one particular job then the other machine can’t process the same job. Different schedule’s order might produce different total processing time. The result of this scheduling problem will be total processing time and schedule’s order. This paper uses clonal selection as the algorithm to solve this problem. The clonal selection algorithm comes from the concept of an artificial immune system. It's developed by copying a human’s immune system behavior. A human’s immune system can differentiate foreign objects and eliminate the objects by creating an antibody. An antibody will go to a cloning process and will mutate to further enhance itself. Clonal selection algorithm applies this cloning and mutation principle to find the most optimal solution. The goal is to find the best schedule’s order and makespan. Taillard’s benchmark is used to verify the quality of the result. To compare the result, we use two values: the upper bound and the lower bound. The upper bound is used to describe the best result of a scheduling problem that has been conducted using a certain environment. On the contrary, the lower bound shows the worst. Experiments on changing the algorithm's parameters are also conducted to measure the quality of the program. The parameters are the number of iterations, mutations, and clone numbers. According to the experiment's results, the higher the number of iteration, mutation rate, and clone number, the better solution for the problem. Clonal selection algorithm has not been able to keep up with upper bound or lower bound values from Taillard’s case. Therefore, parameters need to be increased significantly to increase the chance to produce the optimum result. The higher number of parameters used means the longer time needed to produce the result.

Sign in / Sign up

Export Citation Format

Share Document