Optimization of Petrochemical Assets Using Linear Programming

26 Submissions
251 Views
$20,000 USD
Challenge under evaluation

Challenge overview

OVERVIEW

The goal of this Challenge is to optimize petrochemical assets by using Linear Programming.

We use Linear Programming (LP) software to optimize petrochemical assets at one of our refineries. The software uses certain parameters (prices, feedstock availability, product demands, unit constraints, etc.) to identify the optimal refinery operation and feed-product mix. This optimization also includes predicted time spent producing fuels products and time producing lubricants products.

Unfortunately, LP software has certain limitations in solving multiple equations simultaneously. As a result, attempts to optimize complex operations of a modern-day refinery require extended computing time, making the whole process inefficient.

By posting this Challenge, we want to find a “golden middle”: a Linear Programming modelling approach that can deal with multiple modes of operations and yet remain within reasonable execution times and the capacities of the LP platform.

This is a Prize Challenge which requires a written proposal to be submitted and there will be a guaranteed award for at least one submitted solution. By submitting a proposal, the Solver grants the Seeker a right to use any information included in their proposal.

The total guaranteed payout will be $20,000, with at least one award being no smaller than $5,000 and no award being smaller than $2,500. This will be determined based on theoretical evaluation of the proposals by the Seeker.

Submissions to this Challenge must be received by 11:59 PM (US Eastern Time) on February 12, 2024.

- Login or register your interest to start solving!

Find out more about participation in Wazoku Crowd Challenges.

THE CHALLENGE

Background

We use Linear Programming (LP) software to optimize petrochemical assets at one of our refineries.

The software allows us to represent the refinery, which can operate to make fuels products or lubricant products (two modes of operations). The LP model reflects actual operations of the refinery and can be used for crude oil evaluation, production planning, and scheduling. It uses prices, feedstock availability, product demands, and unit constraints to identify the optimal refinery operation and feed-product mix. This optimization includes establishing the time spent in each mode of operation.

We want to use our existing refinery LP model to optimize the percentage of time in producing fuels products versus producing lubricants products, while considering the actual processing unit and blending constraints for scheduling purposes.

We can compare our challenge to the following simple recipe:

If  and  then

If  does not coincide with , because the modes they represent are sequential and not concurrent, then  needs to be optimized  at a lower value, such that  (  ). Because , it already satisfies the constraints, but  could increase up to  if there is an incentive to do so.

Unfortunately, given the currently available single user computer hardware, there is a balance between the number of LP equations the software can solve simultaneously and the execution time that this takes. Large-scale LP problems can make processing inefficient when used for complex operations of a modern-day refinery.

To overcome these limitations, two approaches have been considered:

  1. To partially replicate the refinery LP model by duplicating the crude oil processing unit, a limited number of downstream processing units, and product blending. While this approach has a low impact on the LP matrix size and computing time, it, unfortunately, doesn’t provide a satisfactory answer to every scheduling scenario. This is due to the time-averaging effect of the units that have not been duplicated, which prevents predicting the actual refinery constraints accurately.
  2. To represent each mode of operation by replicating the whole LP model. For example, one LP model would represent making fuels products for 100% of the time, while the other LP model would do so only for lubricant products. This solution will provide answers to almost every scenario, as all aspects of the refinery LP model will be duplicated. However, this approach would require adding LP structure to optimize time spent by each model. Unfortunately, adding LP equations by replicating the entire model will require significant additional execution time for an LP solver to resolve.

By posting this Challenge, we want to find a “golden middle”: an approach to optimize the Refinery LP model so that it could deal with multiple modes of operations and yet remain within reasonable execution times and the capacities limits of the LP software platform.

SOLUTION REQUIREMENTS

An ideal solution to this Challenge will be an LP model demonstrating time optimization of different modes of operations, with a clear description of the equation structure and explanation of how such a model should be used to achieve the desired objective.

Although our preferred solution would be immediately applicable to optimizing petrochemical assets, we are willing to consider innovative solutions based on other types of asset optimization. In the latter case, any guidance to customize the proposed solution to our needs will be highly appreciated.

For any proposed solution, the accompanying explanation should list any introduced assumptions and be supported by existing examples from industry precedents or publications.

Regardless of the approach, the proposed solution should meet the following Solution Requirements:

  1. The proposed solution should be able to manage a refinery having at least 15 processing units and working in three different modes.
  2. The proposed solution should have a matrix size which is less than replicating the whole model for each mode. Obviously, the lower the level of the increase, the more favorably the solution will be judged.
  3. Any proposal should include a description and/or LP model of the situation before and after the applied solution.
  4. Ideally, the proposed solution should be scalable to handle models of higher complexity (increased numbers of processing units and modes)

At the same time, we will not accept solutions based on:

  • The use of distributed computer networks to overcome single user computing limitations.
  • The removal of time spent in each mode as an optimization variable.

Solutions with Technology Readiness Levels (TRLs) 3-6 are invited.

This is a Prize Challenge, which has the following features:

  1. There will be a guaranteed award of $20,000, with at least one award being no smaller than $5,000 and no award being smaller than $2,500.
  2. The award distribution will be determined after theoretical evaluation of the proposals by the Seeker.
  3. By submitting a proposal, the Solver grants the Seeker a right to use any information included in their proposal.

 

YOUR SUBMISSION

Please login and register your interest, to complete the submission form.

The submitted proposals must be written in English and can include:

  1. Participation type – you will first be asked to inform us how you are participating in this challenge, as a Solver (Individual) or Solver (Organization).
  2. Solution Stage - the Technology Readiness Level (TRL) of your solution, TRL1-3 ideation stage, TRL4-6 proof of concept stage, TRL7-9 production ready stage.
  3. Problem & Opportunity - highlight the innovation in your approach to the Problem, its point of difference, and the specific advantages/benefits this brings (up to 500 words).
  4. Solution Overview - detail the features of your solution and how they address the Solution Requirements (up to 500 words, there is space to add more, and to add any appropriate supporting data, diagrams, etc).
  5. Experience - Expertise, use cases and skills you or your organization have in relation to your proposed solution. The Seeker may wish to partner at the conclusion of the Challenge; please include a statement indicating your interest in partnering (up to 500 words).
  6. Solution Risks - any risks you see with your solution and how you would plan for this (up to 500 words).
  7. Timeline, capability and costs - describe what you think is required to deliver the solution, estimated time and cost (up to 500 words).
  8. References - provide links to any publications or press releases of relevance (up to 500 words).

Wazoku encourages the use by Solvers of AI approaches to help develop their submissions, though any produced solely with generative AI are not of interest.

Find out more about participation in Wazoku Crowd Challenges.

Submissions to this Challenge must be received by 11:59 PM (US Eastern Time) on February 12, 2024.

Late submissions will not be considered.

Your submission will be evaluated by the evaluation team first reviewing the information and content you have submitted at the submission form, with attachments used as additional context to your form submission. Submissions relying solely on attachments will receive less attention from the evaluation team.

After the Challenge submission due date, the Seeker will complete the review process and make a decision with regards to the winning solution(s) according to the timeline in the Challenge header. All Solvers who submit a proposal will be notified about the status of their submissions.

Use the slider to explore how the Challenge process works:

Register

Review & Accept

Submit

Win

To start solving this Challenge, log in to the Challenge Center or register as a Solver

Visit the Wazoku Crowd Community to start the conversation