Presubmission Inquiry for peer reviewing of our Python Package pygulp for weighted goal linear programming

[DrRoomani]

Submitting Author: AIDE LAB (@DrRoomani)
Package Name: pygulp - https://pypi.org/project/pygulp/
One-Line Description of Package: PyGuLP: A Python package for Weighted Goal Linear Programming
Repository Link: https://github.com/aidelab-iitbombay/GLP
EiC: TBD

---

Code of Conduct & Commitment to Maintain Package

• I agree to abide by [pyOpenSci's Code of Conduct][PyOpenSciCodeOfConduct] during the review process and in maintaining my package after should it be accepted.
• I have read and will commit to package maintenance after the review as per the [pyOpenSci Policies Guidelines][Commitment].

Description

Traditional linear programming frameworks are well suited to optimizing a single objective subject to feasibility constraints, but they usually fail in solving problems with conflicting objectives such as providing a protein rich diet at lower costs. Goal Linear Programming (GLP) extends linear programming by allowing targets to be modelled explicitly and by minimizing deviations from those targets. Among GLP variants, Weighted Goal Linear Programming (WGP) offers a transparent and interpretable formulation in which the relative importance of competing targets is encoded through priority weights.
PyGuLP is a lightweight Python package that implements WGP using standard linear programming constructs. Users define decision variables, feasibility constraints, and a set of goals expressed as linear functions with target values. For each goal, the package automatically introduces under- and over-deviation variables and constructs the corresponding goal-linking constraints. The resulting linear program minimizes a weighted sum of deviations and is solved using established linear solvers. PyGuLP makes goal-based optimization accessible to applied researchers while preserving full control over the underlying optimization model.

Community Partnerships

NA

Scope

- [x] Workflow automation

Domain Specific

• Optimization

---

• Our package qualifies as a workflow automation package as it is designed to take goals and feasibility constraints as input and automatically build the full WGP model (deviation variables, goal-linking equations, and the final weighted objective), solve it using standard solvers, and return goal achievement and deviations in a consistent output format that supports repeatable scenario runs and clean comparisons across alternatives. Well suited to fields where optimized allocation of resources is a regular actitvity due to constantly changing inputs, and not just a one time effort.

• The package has broad applications in multi-objective optimization and can be used for resource allocation in the field of nutrition, hospital management, environmental sciences to name a few.

• To the best of our knowledge, there is no python package that addresses Goal Linear Programming. Our package adds functionalities not currently provided under the popular optimization package PuLP.

[kysolvik]

Hi @DrRoomani!

My name is Kylen, I'm a member of the EiC team. Thanks for your presubmission inquiry regarding PyGuLP! I think this could potentially be in scope, but I'm going to consult other members of the editor team. A few things that would help us:

1. Could you respond in a comment with the full presubmission template filled out? The version in this issue is missing some parts. Note that there's no need to delete lines/check-boxes that aren't used.
2. This probably isn't necessary in the presubmission stage, but it would be helpful to have more examples of how this package supports scientific or research activities. Ways to meet this include tutorials/examples of how it is used in scientific workflows or scientific publications that used this package. Currently, the example in the project README shows a single-goal problem, which doesn't fully highlight the added functionality of PyGuLP compared to PuLP.
3. PuLP seems very broadly used for linear programming. Seeing as PyGuLP is built around PuLP, what is the benefit of creating a separate package vs. adding this functionality to PuLP via a pull request? There's currently an open issue requesting this functionality over there. Note that I'm not suggesting that it shouldn't be it's own package, just curious as to your thinking!

Thanks again for your presubmission inquiry, and looking forward to hearing your thoughts!

[DrRoomani]

---

name: Presubmission Inquiry
about: Not sure if your software fits? Submit a presubmission inquiry to the editors
title: ''pygulp" for weighted goal linear programming
labels: presubmission
assignees: ''

---

Submitting Author: Roomani Srivastava (@DrRoomani)
Package Name: pygulp - https://pypi.org/project/pygulp/
One-Line Description of Package: PyGuLP: A Python package for Weighted Goal Linear Programming
Repository Link: https://github.com/aidelab-iitbombay/GLP
EiC: TBD

---

Code of Conduct & Commitment to Maintain Package

• I agree to abide by pyOpenSci's Code of Conduct during the review process and in maintaining my package after should it be accepted.
• I have read and will commit to package maintenance after the review as per the pyOpenSci Policies Guidelines.

Description

• Include a brief paragraph describing what your package does:

Community Partnerships

We partner with communities to support peer review with an additional layer of
checks that satisfy community requirements. If your package fits into an
existing community please check below:

• Astropy: My package adheres to Astropy community standards
• Pangeo: My package adheres to the Pangeo standards listed in the pyOpenSci peer review guidebook

pygulp doesn't fit either of these communities

Scope

• Please indicate which category or categories this package falls under:

  • Data retrieval
  • Data extraction
  • Data processing/munging
  • Data deposition
  • Data validation and testing
  • Data visualization
  • Workflow automation
  • Citation management and bibliometrics
  • Scientific software wrappers
  • Database interoperability

Domain Specific

• Geospatial
• Education

pygulp has functionality across multiple domains where multi-objective optimizations are required.

• Explain how and why the package falls under these categories (briefly, 1-2 sentences). For community partnerships, check also their specific guidelines as documented in the links above. Please note any areas you are unsure of:
  Our package qualifies as a workflow automation package as it is designed to take goals and feasibility constraints as input and automatically build the full WGP model (deviation variables, goal-linking equations, and the final weighted objective), solve it using standard solvers, and return goal achievement and deviations in a consistent output format that supports repeatable scenario runs and clean comparisons across alternatives. Well suited to fields where optimized allocation of resources is a regular actitvity due to constantly changing inputs, and not just a one time effort.

• Who is the target audience and what are the scientific applications of this package?
  The package has broad applications in multi-objective optimization and can be used for resource allocation in the field of nutrition, hospital management, environmental sciences to name a few.

• Are there other Python packages that accomplish similar things? If so, how does yours differ?
  To the best of our knowledge, there is no python package that addresses Goal Linear Programming. Our package adds functionalities not currently provided under the popular optimization package PuLP.
  We are building this as a separate package as we aim to add more functionality in terms of multi-objective optimization as we go.

• Any other questions or issues we should be aware of:

P.S. Have feedback/comments about our review process? Leave a comment here

[DrRoomani]

@kysolvik
Dear Kylen,

Thank you for the response. I have added the full presubmission template in a comment. We will respond to your remaining comments shortly as well.

Thanks,
Roomani

[DrRoomani]

@kysolvik
Dear Kylen. I apologise for the long gap in our response, but here is a reply to your other two comments:

1. Reame has single goal example and showcasing more examples of how pygulp can support other scientific or research activities.
   Response:
   To better illustrate the functionality of pygulp, the linked github repository already contained several worked examples demonstrating multi-goal optimization problems in research contexts. Here is the link: https://github.com/aidelab-iitbombay/GLP/tree/Worked_examples_using_GLP
   These examples include:
   an outpatient resource allocation model balancing turnaround time targets, service coverage, and operational cost,
   a health insurance premium optimization problem balancing profitability, affordability, and coverage targets, and
   a waiting-time target planning problem for outpatient clinics with multiple service-level goals.

These examples demonstrate how PyGuLP supports scientific workflows involving multiple competing targets rather than a single objective function. Based on this feedback, we have now updated the README as well to reference these worked examples and added a concise multi-goal example so that the primary functionality of the package is more immediately visible to users.

2. Why a separate package:

We did consider contributing this functionality directly to PuLP. However, our decision to develop PyGuLP as a separate package was mainly related to scope and long-term evolution. PuLP is designed as a general-purpose linear programming modeler whose core role is to define LP/MIP problems and connect them to different solvers. Goal Programming introduces a different modeling paradigm, where problems are defined in terms of targets and priorities rather than a single objective function.

Implementing goal programming typically requires additional modelling constructs such as automatic deviation variables, goal-linking equations, and mechanisms for managing goal priorities and weighted deviations. These abstractions change how models are specified and solved rather than simply adding another constraint or objective type. Integrating these features directly into PuLP would likely require extending its modelling layer beyond its current scope.

It is nice to know that there is a long-standing interest in adding GP to PuLP. However, we beilieve those discussions have stalled precisely because GP is a "meta-algorithm" rather than a simple mathematical formulation.
By implementing PyGuLP as a separate package built on top of PuLP, we can provide a domain-specific abstraction layer for goal programming while still using PuLP as the underlying LP/MIP modeling engine. This also allows the package to evolve independently as we add more goal-programming classes and variants over time (for example lexicographic, compromise, or range-based goal programming), without increasing the complexity of the PuLP core library.

In this sense, we view PyGuLP as a specialized extension that complements PuLP. PuLP provides the solver interface and LP modeling infrastructure, while PyGuLP provides a higher-level interface for constructing and solving goal-programming models.