Modeling and Optimization: Theory and Applications
  • Slide


Competition

Important dates
Contact
  • For questions about the competition Dr. Karmel S. Shehadeh,
  • For questions about the conference Dr. Albert S. Berahas,
  • For questions about travel & accommodation Sheila Dorney,


13th AIMMS-MOPTA Optimization Modeling Competition

Home Service Assignment, Routing, and Scheduling with Stochastic Service Time, Travel time, and Cancellation

Introduction

   Home services are the provision of providing services to people at their homes, e.g., home health care, home beauty care, banking service, appliance repair service, and more. The home service industry, especially home healthcare, has been rapidly growing worldwide due to emergent changes in family structures, work obligations, aging populations, and outspread of chronic and infectious diseases. In 2018, the global home service market was valued at ∼ $282 Billion and is expected to grow by 18.91% annually from 2019-2026, reaching $1,133.4 Billion by 2026 (Verified Market Research, 2019 [1]). Accordingly, the development of computationally efficient and implementable tools is essential to support decision-making in all areas of the home service industry.
   Home services require professional service teams (or one operator) to travel for delivering the services to geographically distributed customers. Service providers often quote an appointment time (planned service start time) to each customer in advance to avoid delivery failure. Therefore, when home service providers plan for service, they need to decide: how many service teams to hire (i.e., sizing problem), how to assign service teams to the customers (i.e., an assignment problem), how to route service teams (i.e., a vehicle routing problem), and how to assign appointment times for the customers (i.e., appointment scheduling problem). This competition aims to address these specific critical operational decision problems in home service. Stochasticity is an intrinsic property of the home service assignment, routing, and appointment scheduling (H-SARA) problem. Here, we will focus on solving H-SARA under three key random factors: service duration, travel time, and customer cancellation. Your team’s goal is to develop an efficient and implementable method to solve H-SARA, which will be used by home service providers.

[1] Verified Market Research. Global home services market by deployment, by geographic scope and forecast to 2026., Published Date: April 2019.

See the full problem description for more information.
Problem description:  MOPTA2021_Competition.pdf



Sponsor




Registration

You have to register here: registration



Solver Application Download

You are free to use any software of your choice, but it is recommended to use AIMMS for your submission. All source code must be included, properly documented, and results must be reproducible. AIMMS is an industry-leading rapid model building and deployment platform perfected for over 30 years. AIMMS provides an enjoyable and robust way to not only build optimization models but to deploy them as optimization applications to be used by business professionals. You can develop analytical models and highly interactive end-user interfaces all within the same AIMMS environment.

Your team can obtain AIMMS software and request the free academic license from this link.


Submission

All files should be submitted as by the deadline a single email with a single attached zip file to . Please start your submission email's subject line with [MOPTA Competition 2021]. The body of the email should forward the registration confirmation email received at registration time, as the way to authenticate the submitting team. Good luck!

Eligibility

Teams of at most three students can participate. The team leader must be a graduate student, though the other members of the team can be advanced undergraduate students. Each member of the team must be registered as a full-time student at a recognized educational institution during the Spring term of the 2020-2021 Academic Year. Students with any background are eligible. Collaboration between students from different departments is strongly encouraged.

Each team must declare a team advisor with which the team may consult about the problem and their solution. It is the team advisor’s responsibility to ensure that the students have appropriate knowledge for the competition. The team advisor should not be involved as a participant in the solution process.

As the conference is international, so is the competition. Teams from all over the world can participate, as long as at least one team member can come to the conference, should the team make it to the final round.

The official language of the competition is English.

Teams are asked to register to the competition as soon as they start working on the problem. There is a separate deadline for the submission of the solution (see Important dates).

If you have questions about the problem or the competition format in general, please contact Dr. Karmel Shehadeh at . If you have questions about the AIMMS software and licensing related issues, please contact .

Competition Format

The competition consists of a few stages. In the first stage the teams are asked to develop models and solution methods (see the problem description above) and provide an implementation of the optimization models in AIMMS. The teams must submit a complete solution to the problem, including: implementation of the optimization models in AIMMS, ideally already including a graphical user interface that provides the user with graphical and textual output; solutions of the models for the given case study, and for other data sets generated by the teams, if any; a maximum 15 page report that discusses the models developed (along with mathematical background), the solutions obtained, and further recommendations. The AIMMS implementation is recommended for the optimization models only: the teams can use any tool of their choice build the proposed model(s) and/or solution method or run machine learning algorithms on the case study provided if this is a part of their approach (documented source codes should still be included). The teams should also keep in mind that opportunities to improve the solution approach and the interface will be offered in the next phase. A panel of judges, including representatives from both the conference organizing committee and AIMMS evaluates the submissions, provides feedback to the teams and invites finalists to continue in the second stage of the project and present their work at a dedicated session of the conference. In this second stage, the finalists will receive advice from the panel on ways in which they can improve their model and solution and have time before the conference to continue to improve their solution. After the presentations at the conference, the judges will ask questions. The finalists are ranked based on a combined score for their model, implementation, report, solution, oral presentation, and answers to the judges’ questions. The decision of the judges is final and cannot be appealed.

Prizes

  • $1200 for the winning team (e.g., to be used for a new PS5, a set of games, and a great flat party once COVID allows us)
      + certificates for each team member
  • $600 for the second team (e.g., to be used for a new PS5, a few games, and a small group party once COVID allows us)
      + certificates for each team member
  • $300 for the third team (e.g., to be used for a classic mini-game console and a small group party once COVID allows us)
      + certificates for each team member

Copyright

By submitting an entry to the competition you agree that the organizers own the copyright to a copy of your submission. This does not limit your rights to publish your work, give talks, posters, etc., but grants the organizers of the competition the right to use your work.