A Model for Optimal Design of Mixed Renewable Warranty Policy for Non-Repairable Weibull Life Products under Conflict between Customer and Manufacturer Interests
A model is presented to find the optimal design of the
mixed renewable warranty policy for non-repairable Weibull life
products. The optimal design considers the conflict of interests
between the customer and the manufacturer: the customer interests
are longer full rebate coverage period and longer total warranty
coverage period, the manufacturer interests are lower warranty cost
and lower risk. The design factors are full rebate and total warranty
coverage periods. Results showed that mixed policy is better than full
rebate policy in terms of risk and total warranty coverage period in all
of the three bathtub regions. In addition, results showed that linear
policy is better than mixed policy in infant mortality and constant
failure regions while the mixed policy is better than linear policy in
ageing region of the model. Furthermore, the results showed that
using burn-in period for infant mortality products reduces warranty
cost and risk.
Reliability, Mixed warranty policy, Optimization,Weibull Distribution.
A Dual Fitness Function Genetic Algorithm: Application on Deterministic Identical Machine Scheduling
In this paper a genetic algorithm (GA) with dual-fitness function is proposed and applied to solve the deterministic identical machine scheduling problem. The mating fitness function value was used to determine the mating for chromosomes, while the selection fitness function value was used to determine their survivals. The performance of this algorithm was tested on deterministic identical machine scheduling using simulated data. The results obtained from the proposed GA were compared with classical GA and integer programming (IP). Results showed that dual-fitness function GA outperformed the classical single-fitness function GA with statistical significance for large problems and was competitive to IP, particularly when large size problems were used.
Machine scheduling, Genetic algorithms, Due dates, Number of tardy jobs, Number of early jobs, Integer programming, Dual Fitness functions.
Effect of Progressive Type-I Right Censoring on Bayesian Statistical Inference of Simple Step–Stress Acceleration Life Testing Plan under Weibull Life Distribution
This paper discusses the effects of using progressive Type-I right censoring on the design of the Simple Step Accelerated Life testing using Bayesian approach for Weibull life products under the assumption of cumulative exposure model. The optimization criterion used in this paper is to minimize the expected pre-posterior variance of the Pth percentile time of failures. The model variables are the stress changing time and the stress value for the first step. A comparison between the conventional and the progressive Type-I right censoring is provided. The results have shown that the progressive Type-I right censoring reduces the cost of testing on the expense of the test precision when the sample size is small. Moreover, the results have shown that using strong priors or large sample size reduces the sensitivity of the test precision to the censoring proportion. Hence, the progressive Type-I right censoring is recommended in these cases as progressive Type-I right censoring reduces the cost of the test and doesn't affect the precision of the test a lot. Moreover, the results have shown that using direct or indirect priors affects the precision of the test.
Reliability, Accelerated life testing, Cumulative exposure model, Bayesian estimation, Progressive Type-I censoring, Weibull distribution.
Estimating the Life-Distribution Parameters of Weibull-Life PV Systems Utilizing Non-Parametric Analysis
In this paper, a model is proposed to determine the life
distribution parameters of the useful life region for the PV system
utilizing a combination of non-parametric and linear regression
analysis for the failure data of these systems. Results showed that this
method is dependable for analyzing failure time data for such reliable
systems when the data is scarce.
Masking, Bathtub model, reliability, non-parametric
analysis, useful life.
Binary Programming for Manufacturing Material and Manufacturing Process Selection Using Genetic Algorithms
The material selection problem is concerned with the
determination of the right material for a certain product to optimize
certain performance indices in that product such as mass, energy
density, and power-to-weight ratio. This paper is concerned about
optimizing the selection of the manufacturing process along with the
material used in the product under performance indices and
availability constraints. In this paper, the material selection problem
is formulated using binary programming and solved by genetic
algorithm. The objective function of the model is to minimize the
total manufacturing cost under performance indices and material and
manufacturing process availability constraints.
Optimization, Material selection, Process selection,