FURTHER STUDY REPAIR-REPLACEMENT STRATEGY IN TWO DIMENTIONAL WARRANTY

This thesis presents analysis about one of warranty servicing strategies, which is replace replacement strategy given the usage rate in the form of an unknown distribution. The previous research related with this strategy talks about how to apply repair-replacement <br /> <br /> <...

Full description

Saved in:
Bibliographic Details
Main Author: (NIM: 20812003); Pembimbing : Sapto Wahyu Indratno, Ph.D. , RONNY
Format: Theses
Language:Indonesia
Online Access:https://digilib.itb.ac.id/gdl/view/20481
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Institut Teknologi Bandung
Language: Indonesia
Description
Summary:This thesis presents analysis about one of warranty servicing strategies, which is replace replacement strategy given the usage rate in the form of an unknown distribution. The previous research related with this strategy talks about how to apply repair-replacement <br /> <br /> <br /> <br /> strategy given a certain value of usage rate. Meanwhile in the field when the manufacturer provides warranty services to their customers, they will find that each customer will have a different kind of usage. Some of them have light usage, while the others have normal or <br /> <br /> <br /> <br /> maybe heavy usage that will affect the failure time of the product. In this thesis the product assumed is vehicle and the purpose is to find the interval for the strategy to meet the smallest warranty cost. <br /> <br /> <br /> <br /> In this thesis, there will be two methods used, which are Monte Carlo method and two dimension kernel. The kernel functions used are Gaussian kernel and triangle kernel. The kernel function can be used to determine the relation between random variable usage rate and failure time and also the joint function of both variables. After the simulation, it will be found out that the kernel function will give a better result than the Monte Carlo method, which is seen from the cost reduction that can be saved by the manufacturer. The other advantage is <br /> <br /> <br /> <br /> that by using kernel function, we can see the plot of the joint function between the two random variables.