Sequential Change Point Detection in Dynamic Non-Stationary Manufacturing Processes

Description of Software and Data

The code and data in the zip file referenced above are a snapshot of the software and data that were used in the research reported in the paper "Sequential Change Point Detection in Dynamic Non-Stationary Manufacturing Processes" by Yuhan Tian, Abolfazl Safikhani and Kamran Paynabar.

The goal of this repository is to replicate the numerical experiments in the paper.

Computer and Software Environment

The following describes the computer hardware conditions and software environment on which the authors produce the results reported in the paper.

Numerical experiments were conducted using R 4.5.0 on a computer equipped with a 12th Gen Intel(R) Core(TM) i7-12700H (2.30 GHz) CPU and 32 GB of RAM.

Dependencies

The code in this repository requires the following dependencies. The dependency version number corresponds to the version of the package with which the code was tested.

• sparsevar 1.0.0
• ggplot2 3.5.2
• nnls 1.6
• MASS 7.3-65
• blockmatrix 1.0
• expm 1.0-0
• SimDesign 2.19.2
• doParallel 1.0.1
• doRNG 1.8.6.3
• patchwork 1.3.0
• ocp 0.1.1
• Hotelling 1.0-8
• ocd 1.1

Installation

The code and data are provided as a single compressed (.zip) archive. After downloading and extracting the archive, the directory contains the following files: Code_Figure3.R, Code_Figure4b.R, Code_Figure4d.R, and Code_Figure5.R, which reproduce the numerical experiments and figures in the paper; Functions.R, which contains the functions used by the scripts; and SemiData.csv, which contains the data used in the semiconductor experiment. Before running the scripts, please set the working directory in R to the folder containing these files (for example, using setwd("path_to_the_folder")). The required R packages can be installed from CRAN using the following command:

install.packages(c("sparsevar","ggplot2","nnls","MASS","blockmatrix", "expm","SimDesign","doParallel","doRNG", "patchwork","ocp","Hotelling","ocd"))

After installing the required packages and setting the working directory, the scripts can be run directly in R to reproduce the results reported in the paper.

Reproducibility Workflow

To reproduce the results in Figure 3 in the paper

• Data Files: Simulated internally by the code
• Code Files: Code_Figure3.R
  
• Output: Visualization of a simulated dataset and the detected change points, along with the average increase in false alarms reported in Section 2
• Run Time at the Above-Specified Computer Conditions: 10 minutes

To reproduce the results in Figure 4b in the paper

• Data File: Simulated internally by the code
• Code File: Code_Figure4b.R
• Output: Average detection delays of the methods (excluding TSL and STSSD) for different jump sizes.
• Run Time at the Above-Specified Computer Conditions: 1.5 hours

To reproduce the results in Figure 4d in the paper

• Data File: Simulated internally by the code
• Code File: Code_Figure4d.R
• Output: Average detection delays of the methods (excluding TSL and STSSD) under different sparsity levels.
• Run Time at the Above-Specified Computer Conditions: 1.5 hours

To reproduce the results in Figure 5 in the paper

• Data File: SemiData.csv
• Code File: Code_Figure5.R
• Output: Visualization of the data and detected change points, as well as the model parameter estimates and shift estimates in Section 8.2.
• Run Time at the Above-Specified Computer Conditions: 1 minute

Cite

To cite the contents of this repository, please cite both the paper and this repository using their respective DOIs.

Article: https://doi.org/10.1287/ijds.2025.0068
Software and Data Repository: https://doi.org/10.1287/ijds.2025.0068.cd

License

Copyright (c) (2026 Tian, Safikhani, Paynabar)

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.