Online Modeling and Monitoring for Dependent Dynamic Processes Under Resource Constraints

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 "Online Modeling and Monitoring for Dependent Dynamic Processes Under Resource Constraints" by Tanapol Kosolwattana, Huazheng Wang, and Ying Lin. This repository is also available via GitHub.

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.

Experiments were conducted on the Sabine high-performance computing cluster at the University of Houston, using nodes equipped with Intel® Xeon® processors and 28 CPU cores per job allocation across two nodes, with 28 GB of available memory.

Python 3.10.12 and R 4.3 were used for all analysis and simulations.

Dependencies

• Matplotlib 3.8.2
• NumPy 1.26.2
• pandas 2.1.3
• SciPy 1.11.4
• seaborn 0.13.0

Installation

• Install Python 3.10.12 and RStudio 2023.3.1.446 with R 4.3
• Create a virtual environment and activate it.
• Upgrade pip: python -m pip install --upgrade pip
• Install dependencies: python -m pip install -r requirements.txt
• Run experiments from the project root directory which includes the codes: python ./code/FILE NAME.py. For R files, open the files and run in RStudio.
• Data files are located at the Data folder (Please download folders from the provided Google Drive).

File Structure

• Analysis scripts dare in the code/ folder.
• Script data and outputs are saved in results/ folder.
• Core implementation of methods in the paper is in the code/ folder.

Reproducibility Workflow

To reproduce the results in Figure 2 in the paper

• Data Files:
  Y(noise0.1).csv
Beta(noise0.1).csv
C(noise0.1).csv
Q(noise0.1).csv
W(noise0.1).csv
c_init4.csv
q_init4.csv
• Code Files:
  C129.py
C163.py
• Output: The four plots in Figure 2 as four png files
• Run Time at the Above-Specified Computer Conditions: ~30 hours

To reproduce the results in Figure 3 in the paper

• Data Files:
  Y(noise0.1).csv
Beta(noise0.1).csv
C(noise0.1).csv
Q(noise0.1).csv
W(noise0.1).csv
c_init4.csv
q_init4.csv
• Code Files:
  run_miss_k.py
• Output: The plot in Figure 3 as one png file
• Run Time at the Above-Specified Computer Conditions: ~24 hours

To reproduce the results in Table 3 in the paper

• Data Files:
  Y(noise0.1).csv
Beta(noise0.1).csv
C(noise0.1).csv
Q(noise0.1).csv
W(noise0.1).csv
c_init4.csv
q_init4.csv
• Code Files:
  run_G.py
• Output: The value of cumulative regrets under different G values stated in Table 3 of Section 4.2
• Run Time at the Above-Specified Computer Conditions: ~57 hours

To reproduce the results in Figure 4 in the paper

• Data Files:
  small_sig_beta.csv
small_sig_Y.csv
small_sig_W.csv
c_init4.csv
q_init4.csv
• Code File: run_sigma.py
• Output: The plot in Figure 4 as one png file
• Run Time at the Above-Specified Computer Conditions: ~12 hours

To reproduce the results in Figure 5 in the paper

• Data File: For each N = 40, 80, 100, 120, there are 5 sets of data where each set includes the following files:
  set(#number)_(N)_W(noise1).csv
set(#number)_(N)_Beta(noise1).csv
set(#number)_(N)_Q(noise1).csv
set(#number)_(N)_C(noise1).csv
set(#number)_(N)_Y(noise1).csv
set(#number)_N_q.csv
set(#number)_N_c.csv
• Code Files:
  diffN.R
C129.py
• Output: The plot in Figure 5 as one png file
• Run Time at the Above-Specified Computer Conditions: ~50 hours

To reproduce the results in Figure 6 in the paper

• Data File: For each K = 3, 5, there are 5 sets of data where each set includes the following files:
  Set(#number)_Beta(K5).csv
Set_(#number)_W(K5).csv
Set_(#number)_Y(K5).csv
• Code Files:
  diffN.R
C129.py
• Output: The plot in Figure 6 as one png file
• Run Time at the Above-Specified Computer Conditions: ~20 hours

To reproduce the results in Figure 7 in the paper

• Data File:
  poly2_Y_all(mon).csv
poly2_W.csv
• Code Files:
  C233.py
C266.py
• Output: The two plots in Figure 7 as two png files
• Run Time at the Above-Specified Computer Conditions: ~30 hours

To reproduce the results in Figure 8 in the paper

• Data File:
  Y250.csv
poly2_W.csv
• Code Files:
  C233.py
C266.py
• Output: The two plots in Figure 8 as two png files
• Run Time at the Above-Specified Computer Conditions: ~3 hours

To reproduce the results in Figure 1 in Appendix D

• Data File:
  Y(noise0.1).csv
Beta(noise0.1).csv
C(noise0.1).csv
Q(noise0.1).csv
W(noise0.1).csv
c_init4.csv
q_init4.csv
• Code Files:
  run_comp_conv.py
• Output: The plot in Figure 1 as one png file
• Run Time at the Above-Specified Computer Conditions: ~12 hours

To reproduce the results in Figure 2 in Appendix D

• Data File:
  Beta_hk.csv
Y_hk.csv
A_hk.csv
c_init4.csv
q_init4.csv
• Code Files:
  run_hk.py
• Output: The plot in Figure 2 as one png file
• Run Time at the Above-Specified Computer Conditions: ~12 hours

To reproduce the results in Figure 3 in Appendix D

• Data File:
  Y(noise0.1).csv
Beta(noise0.1).csv
C(noise0.1).csv
Q(noise0.1).csv
W(noise0.1).csv
c_init4.csv
q_init4.csv
• Code Files:
  run_predy.py
• Output: The plot in Figure 3 as one png file
• Run Time at the Above-Specified Computer Conditions: ~12 hours

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.2023.0021
Software and Data Repository: https://doi.org/10.1287/ijds.2023.0021.cd

License

Copyright (c) (2025 Tanapol Kosolwattana)

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.