Abstract

The Cumulative Sum (CUSUM) chart is widely used and has many applications in different fields such as finance, medical, engineering, and other fields. In real applications, there are many situations in which the observations of random processes are serially correlated, such as a hospital admission in the medical field, a share price in the economic field, or a daily rainfall in the environmental field. The common characteristic of control charts that has been used to evaluate the performance of control charts is the Average Run Length (ARL). The primary goals of this paper are to derive the explicit formula and develop the numerical integral equation of the ARL for the CUSUM chart when observations are seasonal autoregressive models with exogenous variable, SARX(P,r)<sub>L</sub> with exponential white noise. The Fredholm Integral Equation has been used for solving the explicit formula of ARL, and we used numerical methods including the Midpoint rule, the Trapezoidal rule, the Simpson's rule, and the Gaussian rule to approximate the numerical integral equation of ARL. The uniqueness of solutions is guaranteed by using Banach's Fixed Point Theorem. In addition, the proposed explicit formula was compared with their numerical methods in terms of the absolute percentage difference to verify the accuracy of the ARL results and the computational time (CPU). The results obtained indicate that the ARL from the explicit formula is close to the numerical integral equation with an absolute percentage difference of less than 1%. We found an excellent agreement between the explicit formulas and the numerical integral equation solutions. An important conclusion of this study was that the explicit formulas outperformed the numerical integral equation methods in terms of CPU time. Consequently, the proposed explicit formulas and the numerical integral equation have been the alternative methods for finding the ARL of the CUSUM control chart and would be of use in fields like biology, engineering, physics, medical, and social sciences, among others.

Highlights

  • The main intention of statistical process control (SPC) is to provide a technique for improving productivity

  • The explicit formula is compared with numerical integral equation (NIE) methods using the midpoint rule with 500 nodes to verify which method is better with the absolute percentage difference and the computational time (CPU) times

  • The absolute percentage difference, Diff(%), and the CPU time are used to compare the performance of the explicit formula and the NIE methods for the Cumulative Sum (CUSUM) chart

Read more

Summary

Introduction

The main intention of statistical process control (SPC) is to provide a technique for improving productivity. Control charts are one of the efficient tools of SPC for detecting changes in the mean or variations in the process. The SPC charts, such as the Shewhart chart, were first introduced by Shewhart [1], the Cumulative Sum (CUSUM) chart, introduced by Page [2], and the Exponentially Weighted Moving Average (EWMA) chart, proposed by Roberts [3]. These are used to monitor product quality and to detect the occurrence of special causes that may be indicative of out-of-control situations. The Mathematics and Statistics 10(1): 88-99, 2022 purpose of this paper is to study the Fredholm integral equations method to derive a closed-form solution of the Average Run Length for the CUSUM chart when observations are seasonal autoregressive with an exogenous variable; SARX(P,r)L model with exponential distribution white noise

Objectives
Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call