SPATIAL SIMULATION OF SEIR EPIDEMIC MODEL USING CELLULAR AUTOMATA

• Main Author: Zaki Almuzakki, Muhammad
• Format: Theses
• Language: Indonesia
• Subjects: Ilmu alam dan matematika
• Online Access: https://digilib.itb.ac.id/gdl/view/37369
• Institution: Institut Teknologi Bandung

Severe acute respiratory syndrome (SARS) is one of diseases that quickly spread through populations and has relatively long infection period. It has been shown that almost fifty percent of susceptible individuals that have been in contact with the infected individuals are also found to be infected by SARS. At the same time, the relatively long infection period will cause the speed of the spread increases each day. Therefore, good analysis of SARS becomes an important matter for making the best decision. According to WHO, mathematics model is one of useful tools in analyzing a spreading disease and making the best decision for it. There are many mathematics models built using a system of ordinary differential equations and its improvements. The combination between SEIR epidemic model and cellular automata (CA) called SEIR-CA model is used in this research to analyze the spread of SARS in Hong Kong at 2003. This phenomenon is chosen because of its properties are suitable with the assumption given in the model, that is, it has incubation period, it spreads through both direct and indirect contact between individuals, and has not been affected by other parameters such as precautions like vaccination. The combination of SEIR epidemic model and CA enables the disease to spread from one to another region through indirect contacts that are given by transition mechanism in CA’s neighborhood system. An improvement to the parameter of neighborhood’s impact of the previously established SEIR-CA model is proposed in this research. The simulation results of the improved SEIR-CA model show that the model is well established for both population with uniformly distributed individuals and randomly distributed individuals. The improved SEIR-CA model is then tested with the parameters obtained from thedisease spread data in Hong Kong, 2003, on the population with the distribution of the individuals is randomly generated. The simulation result shows that the graph has similar trend with the real data. On the other hands, the graph of the simulation results more looks like the expanded version of the real data. In addition to the test of the model on the spread of SARS in Hong Kong, 2003, several control mechanisms in the form of vaccination are also proposed in this research. The vaccine is given in a constant ratio for each day with several vaccination period. The simulation results show that the longer the vaccination period gives greater impact to the reduction of the number of infected individuals. Furthermore, the faster the decision of vaccination is taken can prevent the disease to spread to other regions.