Submission: Group 17: Analysis and Prediction of Airline Delays in 2019

Submitting authors: Siddharth Balodi, Charles Benkard, Mikel Ibarra Gallardo, and Stephanie Ta

Repository: https://github.com/DSCI-310-2024/dsci-310_airline-delay-classification_group-17/releases

Abstract/executive summary:

In this project, we delve into a 2019 airline delays dataset to dissect the intricate factors contributing to flight disruptions. Our analysis aims to address pivotal questions, such as what are the primary drivers behind flight delays and cancellations? Are certain airlines or airports more vulnerable to these disruptions? How do external factors like adverse weather conditions or air traffic congestion exacerbate flight schedules? While previous studies have touched upon aspects of flight disruptions, we endeavor to provide a deeper understanding through the lens of analytical techniques, including descriptive statistics, data visualization, and machine learning algorithms (specifically KNN classification).

Through our exploration, we have unearthed compelling insights. We have found that airport congestion, inclement weather, and airline operational issues are key contributors to flight disruptions. Furthermore, our analysis has revealed disparities in performance among airlines and airports, shedding light on areas ripe for operational enhancements and service improvements. By leveraging these findings, stakeholders within the aviation industry can make informed decisions aimed at minimizing disruptions and enhancing overall operational efficiency.

The significance of our research extends beyond the realm of academia. By unraveling the complexities of flight disruptions, we aim to empower decision-makers with actionable insights to navigate the challenges inherent in air travel. Moreover, our findings hold the potential to drive positive change within the aviation industry, ultimately leading to a more seamless and reliable travel experience for passengers worldwide.

Editor: @ttimbers

Reviewer: Yanxin Liang, Rashi Selarka, Caden Chan, Zackarya Hamza

[ ] I agree to abide by DSCI 310's Code of Conduct during the review process.

Data analysis review checklist

Reviewer: @cadenmchan

Conflict of interest

[Y] As the reviewer I confirm that I have no conflicts of interest for me to review this work.

Code of Conduct

[Y] I confirm that I read and will adhere to the MDS code of conduct.

General checks

[Y] Repository: Is the source code for this data analysis available? Is the repository well organized and easy to navigate?
[Y] License: Does the repository contain a plain-text LICENSE file with the contents of an OSI approved software license?

Documentation

[Y] Installation instructions: Is there a clearly stated list of dependencies?
[Y] Example usage: Do the authors include examples of how to use the software to reproduce the data analysis?
[Y] Functionality documentation: Is the core functionality of the data analysis software documented to a satisfactory level?
[N] Community guidelines: Are there clear guidelines for third parties wishing to 1) Contribute to the software 2) Report issues or problems with the software 3) Seek support

Code quality

[Y] Readability: Are scripts, functions, objects, etc., well named? Is it relatively easy to understand the code?
[Y] Style guidelides: Does the code adhere to well known language style guides?
[Y] Modularity: Is the code suitably abstracted into scripts and functions?
[Y] Tests: Are there automated tests or manual steps described so that the function of the software can be verified? Are they of sufficient quality to ensure software robsutness?

Reproducibility

[N] Data: Is the raw data archived somewhere? Is it accessible?
[Y] Computational methods: Is all the source code required for the data analysis available?
[Y] Conditions: Is there a record of the necessary conditions (software dependencies) needed to reproduce the analysis? Does there exist an easy way to obtain the computational environment needed to reproduce the analysis?
[N] Automation: Can someone other than the authors easily reproduce the entire data analysis?

Analysis report

[Y] Authors: Does the report include a list of authors with their affiliations?
[Y] What is the question: Do the authors clearly state the research question being asked?
[Y] Importance: Do the authors clearly state the importance for this research question?
[Y] Background: Do the authors provide sufficient background information so that readers can understand the report?
[Y] Methods: Do the authors clearly describe and justify the methodology used in the data analysis? Do the authors communicate any assumptions or limitations of their methodologies?
[Y] Results: Do the authors clearly communicate their findings through writing, tables and figures?
[Y] Conclusions: Are the conclusions presented by the authors correct?
[Y] References: Do all archival references that should have a DOI list one (e.g., papers, datasets, software)?
[Y] Writing quality: Is the writing of good quality, concise, engaging?

Estimated hours spent reviewing: 3

Review Comments:

Project Strengths

The overall file structure and organization of project repository was easy to navigate
The scripts were easy to follow and read; a good level of abstraction in my opinion
Report purpose, methodology, and justification were all very clear and well executed
Analysis conducted made sense and all steps were well-defined and necessary
Usage documentation and dependencies were straightforward and easy to understand
All tests ran without failure

Areas of Improvement

Some file naming conventions were confusing/inconsistent (in src only some scripts have numeric prefixes, in tests only some scripts have test_ prefix)
Some inconsistency with some function abstractions (ex. filter-raw-data.py has definition for functions read() and filter_columns(), but the read.py and filter_columns.py scripts also define functions read() and filter_columns. filter-raw-data.py does not import functions from either of the other scripts.)
requirements.txt is empty file in root directory
Link to original data files are missing/difficult to find, with raw data files not uploaded to github repository (only data/processed) and scripts do not read data from web URL
make all doesn't work because there are no raw data files to access

Building/running Docker container based on usage guidelines didn't work:

docker build -t airline-project

returned the following error:


ERROR: "docker buildx build" requires exactly 1 argument.
See 'docker buildx build --help'.

Usage: docker buildx build [OPTIONS] PATH | URL | -

Start a build

I think the proper build command is just missing a dot:

docker build -t airline-project .


- Building and running Docker container does not install required packages (still have to create env from yml file). I think this is because `requirements.txt` is empty
- `LICENSE` does not have team/team members as authors (DSCI 310 attributed instead)
- Perhaps could add clear guidelines for third parties wishing to 1) Contribute to the software 2) Report issues or problems with the software 3) Seek support

#### Attribution

This was derived from the [JOSE review checklist](https://openjournals.readthedocs.io/en/jose/review_checklist.html) and the ROpenSci review checklist.

Data analysis review checklist

Reviewer: Yuji03b

Conflict of interest

[x] As the reviewer I confirm that I have no conflicts of interest for me to review this work.

Code of Conduct

[x] I confirm that I read and will adhere to the MDS code of conduct.

General checks

[x] Repository: Is the source code for this data analysis available? Is the repository well organized and easy to navigate?
[x] License: Does the repository contain a plain-text LICENSE file with the contents of an OSI approved software license?

Documentation

[x] Installation instructions: Is there a clearly stated list of dependencies?
[x] Example usage: Do the authors include examples of how to use the software to reproduce the data analysis?
[x] Functionality documentation: Is the core functionality of the data analysis software documented to a satisfactory level?
[x] Community guidelines: Are there clear guidelines for third parties wishing to 1) Contribute to the software 2) Report issues or problems with the software 3) Seek support

Code quality

[x] Readability: Are scripts, functions, objects, etc., well named? Is it relatively easy to understand the code?
[x] Style guidelides: Does the code adhere to well known language style guides?
[x] Modularity: Is the code suitably abstracted into scripts and functions?
[ ] Tests: Are there automated tests or manual steps described so that the function of the software can be verified? Are they of sufficient quality to ensure software robsutness?

Reproducibility

[ ] Data: Is the raw data archived somewhere? Is it accessible?
[x] Computational methods: Is all the source code required for the data analysis available?
[x] Conditions: Is there a record of the necessary conditions (software dependencies) needed to reproduce the analysis? Does there exist an easy way to obtain the computational environment needed to reproduce the analysis?
[ ] Automation: Can someone other than the authors easily reproduce the entire data analysis?

Analysis report

[x] Authors: Does the report include a list of authors with their affiliations?
[x] What is the question: Do the authors clearly state the research question being asked?
[x] Importance: Do the authors clearly state the importance for this research question?
[x] Background: Do the authors provide sufficient background information so that readers can understand the report?
[x] Methods: Do the authors clearly describe and justify the methodology used in the data analysis? Do the authors communicate any assumptions or limitations of their methodologies?
[x] Results: Do the authors clearly communicate their findings through writing, tables and figures?
[x] Conclusions: Are the conclusions presented by the authors correct?
[ ] References: Do all archival references that should have a DOI list one (e.g., papers, datasets, software)?
[x] Writing quality: Is the writing of good quality, concise, engaging?

Estimated hours spent reviewing: 2

Review Comments:

Strengths:

The report and CONTRIBUTION.md are highly detailed and straightforward to follow.
The plots in the report are very clear and visually appealing.

Improvements:

The Docker build and run commands don't work.
Despite clear instructions in the README, the example usages in the Makefile are misleading because it lists "make clean" which does not exist.
It would be beneficial to include some descriptions of automated tests or manual steps in the README.
The reference to the paper is missing in the report's HTML file.

Attribution

This was derived from the JOSE review checklist and the ROpenSci review checklist.

Data analysis review checklist

Reviewer: @rashiselarka

Conflict of interest

[x] As the reviewer I confirm that I have no conflicts of interest for me to review this work.

Code of Conduct

[x] I confirm that I read and will adhere to the MDS code of conduct.

General checks

[x] Repository: Is the source code for this data analysis available? Is the repository well organized and easy to navigate?
[x] License: Does the repository contain a plain-text LICENSE file with the contents of an OSI approved software license?

Documentation

[x] Installation instructions: Is there a clearly stated list of dependencies?
[x] Example usage: Do the authors include examples of how to use the software to reproduce the data analysis?
[x] Functionality documentation: Is the core functionality of the data analysis software documented to a satisfactory level?
[ ] Community guidelines: Are there clear guidelines for third parties wishing to 1) Contribute to the software 2) Report issues or problems with the software 3) Seek support

Code quality

[x] Readability: Are scripts, functions, objects, etc., well named? Is it relatively easy to understand the code?
[x] Style guidelides: Does the code adhere to well known language style guides?
[x] Modularity: Is the code suitably abstracted into scripts and functions?
[x] Tests: Are there automated tests or manual steps described so that the function of the software can be verified? Are they of sufficient quality to ensure software robsutness?

Reproducibility

[ ] Data: Is the raw data archived somewhere? Is it accessible?
[x] Computational methods: Is all the source code required for the data analysis available?
[x] Conditions: Is there a record of the necessary conditions (software dependencies) needed to reproduce the analysis? Does there exist an easy way to obtain the computational environment needed to reproduce the analysis?
[ ] Automation: Can someone other than the authors easily reproduce the entire data analysis?

Analysis report

[x] Authors: Does the report include a list of authors with their affiliations?
[x] What is the question: Do the authors clearly state the research question being asked?
[x] Importance: Do the authors clearly state the importance for this research question?
[x] Background: Do the authors provide sufficient background information so that readers can understand the report?
[x] Methods: Do the authors clearly describe and justify the methodology used in the data analysis? Do the authors communicate any assumptions or limitations of their methodologies?
[x] Results: Do the authors clearly communicate their findings through writing, tables and figures?
[x] Conclusions: Are the conclusions presented by the authors correct?
[ ] References: Do all archival references that should have a DOI list one (e.g., papers, datasets, software)?
[x] Writing quality: Is the writing of good quality, concise, engaging?

Estimated hours spent reviewing:

Review Comments:

Strengths:

It's a well organized repository - it's nice that the scripts are numbered as it makes the workflow easy to follow.
The report is also well organized, flows well, and the visualizations are well sized and clear.
The interactive visualizations are a nice touch!
All the docstrings are verbose and very clear - great job.
The environment is easy to install - the dependencies are straightforward and version pinned.

Areas of Improvement:

The scripts should be in a separate scripts directory and the functions should be in src.
The docker build command on the README.md doesn't work - invokes an argument error. It's missing a period (".") at the end. It's also a good idea to work with a docker-compose file instead - makes the starting and closing of containers easier.
The Dockerfile is very unlike the way we've learnt to build it in class.
requirements.txt is empty.
make all command does not work: The Makefile is missing the commands to load the above data files. The raw data can't be found in general (the link or the csv file) so essential to make that fix.
Error in 03_test_split_data.py - there is no import for sys even though it's being used.
There's only 1 reference in the report to the source of the data - there should be more academic or otherwise (e.g. news articles) sources that should be used to corroborate statements being made in the summary and introduction.
Figure 6 on the report is a bit unclear - can't tell what it's trying to convey.
The summary output could be converted into a table to make it look nicer.
The CONTRIBUTING should be more detailed - refer to an example repo from the assignment or another group's file to see what the file should look like.

Attribution

This was derived from the JOSE review checklist and the ROpenSci review checklist.

Data analysis review checklist

Reviewer: zackhamza01

Conflict of interest

[x] As the reviewer I confirm that I have no conflicts of interest for me to review this work.

Code of Conduct

[x] I confirm that I read and will adhere to the MDS code of conduct.

General checks

[x] Repository: Is the source code for this data analysis available? Is the repository well organized and easy to navigate?
[x] License: Does the repository contain a plain-text LICENSE file with the contents of an OSI approved software license?

Documentation

[x] Installation instructions: Is there a clearly stated list of dependencies?
[x] Example usage: Do the authors include examples of how to use the software to reproduce the data analysis?
[x] Functionality documentation: Is the core functionality of the data analysis software documented to a satisfactory level?
[x] Community guidelines: Are there clear guidelines for third parties wishing to 1) Contribute to the software 2) Report issues or problems with the software 3) Seek support

Code quality

[x] Readability: Are scripts, functions, objects, etc., well named? Is it relatively easy to understand the code?
[x] Style guidelides: Does the code adhere to well known language style guides?
[ ] Modularity: Is the code suitably abstracted into scripts and functions?
[x] Tests: Are there automated tests or manual steps described so that the function of the software can be verified? Are they of sufficient quality to ensure software robsutness?

Reproducibility

[x] Data: Is the raw data archived somewhere? Is it accessible?
[x] Computational methods: Is all the source code required for the data analysis available?
[x] Conditions: Is there a record of the necessary conditions (software dependencies) needed to reproduce the analysis? Does there exist an easy way to obtain the computational environment needed to reproduce the analysis?
[x] Automation: Can someone other than the authors easily reproduce the entire data analysis?

Analysis report

[x] Authors: Does the report include a list of authors with their affiliations?
[x] What is the question: Do the authors clearly state the research question being asked?
[x] Importance: Do the authors clearly state the importance for this research question?
[x] Background: Do the authors provide sufficient background information so that readers can understand the report?
[x] Methods: Do the authors clearly describe and justify the methodology used in the data analysis? Do the authors communicate any assumptions or limitations of their methodologies?
[x] Results: Do the authors clearly communicate their findings through writing, tables and figures?
[x] Conclusions: Are the conclusions presented by the authors correct?
[x] References: Do all archival references that should have a DOI list one (e.g., papers, datasets, software)?
[x] Writing quality: Is the writing of good quality, concise, engaging?

Estimated hours spent reviewing:

30 mins

Review Comments:

Please provide more detailed feedback here on what was done particularly well, and what could be improved. It is especially important to elaborate on items that you were not able to check off in the list above.

When viewing the .ipynb analysis file, there are errors in the Methods and Results section, like a ModuleNotFound error and a NameError. (unless you plan on removing .ipynb analysis and just leaving the others?)
I really like how you abstracted a lot of the code from the scripts into functions. Very impressive! What I would suggest is implementing some modularity in your src/ folder, like separating your scripts and functions into two sub folders scripts/ and functions/
Otherwise, I really can't find any fault with your analysis. Your group did a great job! Just a small amelioration to consider, maybe consider rendering a .pdf version in your reports folder like how we did during the Quarto exercise. This is just to give more flexibility, but keep in mind this is not required.
Attribution

This was derived from the JOSE review checklist and the ROpenSci review checklist.

DSCI-310-2024 / data-analysis-review-2024