Freestyle O.R. Supreme Hackathon

Freestyle O.R. Supreme Hackathon logo

The 2019 INFORMS Business Analytics Conference will host a Freestyle OR Supreme Data Hackathon event co-sponsored by Analytics2Go and the INFORMS Data Mining section. The audience for this hackathon is students in the fields that fall under the large umbrella of analytics and business, as well as early career professionals (less than 5 years post their analytics-related degree). The expectations would be to design and deliver an analytics solution with an open-source tool/software component to it, and sell how this solution could be operable and scalable to support the business problem and their customers.

Benefits & Awards

The top four selected teams will be invited to present their solution at the conference. Those teams may receive a conference registration discounted team rate of $1,070 per team member. The following awards will be provided after the presentations are complete:

  • First place team            $2000 + plaque & certificate
  • Second place team      plaque & certificate
  • Third place team          plaque & certificate

Timeline & Key Dates

  • February 25

    Challenge Begins

  • March 15

    Submissions Due

    Initial solution write-up + codes submission submitted by 11:59pm EST

  • March 25

    Team Selections Made

    Top 4 teams speaker invitations will be sent out

  • April 1

    Confirm Participation

    Notify organizing committee of participation and show registration (at least one person from each team required)

  • April 15-17

    Competition Presentations

    Present at scheduled session in front of judging panel

  • April 17

    Top 3 Teams Announced

Eligibility & Team Formation

Teams consisting of 2 to 5 members may be formed. Participants can participate on only one team. The audience for this hackathon are students in fields that fall under the large umbrella of analytics and business, as well as early career professionals (less than 5 years post their analytics-related degree). If you do not fall into one of those two groups, then you are not eligible to participate in this competition. Students must compete with students from their currently enrolled university only. Industry participants must come from the same company. Employees of Analytics2Go are not eligible to participate.

Submissions

There is no submission template, however, following the INFORMS CAP framework is highly encouraged. Submissions are due March 15th by 11:59pm EST to lanhamm@purdue.edu. Submissions received after that time will not be considered. In the submission email, you should have the following information like so:

  • Subject: Attention: INFORMS BA Hackathon Submission
  • Body:
    • Team Name: A chosen team name that does not identify which school or company you are coming from. Be creative.
    • Affiliated school or company:
    • First and last names of all team members along with emails. Please provide phone numbers (if available). The first person listed should be the team captain or primary contact person.
  • Attachments or Link: Email attachments must be less than 7MB in size. Thus, you may put in the body of your email in lieu of attached documents, a link to a public Google Drive or Dropbox folder.
    • Your write-up can be in a Microsoft Word or PDF file. The write-up should only have your team name and not other identifiable author info. Only the submission email will have your names and contact info. Also you should not have codes or screenshots of codes. Rather clearly and concisely indicate in your write-up which files do what.
    • Any codes to reproduce the results should be clearly labeled and described concisely in your report. If there are multiple files, clearly show with a diagram how they relate to each other.

Once you have submitted your solution, you will receive a confirmation email within 24 hours. If you do not receive the confirmation email, email lanhamm@purdue.edu immediately.

Invited Presentations

Once all submissions have been submitted by March 15, a team of internal judges will review the proposed solutions and the top four teams will be selected. Once invitations have been submitted, one member from each team must let the organizing committee know that they will or will not attend the conference to present their team’s final solution by April 1st. When doing this, you must show that you have registered for the conference. Any team that does not have at least one person registered to attend and present their solution will forfeit their invitation and alternate teams will be invited to fill their spot. There will be a couple presentation sessions for the top four teams, one containing two presenters and one containing the remaining two. Teams may have one presenter or it could be a team effort with multiple people discussing certain parts of the solution. However, keep in mind that each scheduled session is only 55 minutes long, and each talk is expected to last 20 minutes total to allow for some Q&A at the end.

Competition Judging

Judges will be focused on the following when evaluating solutions and invited presentations:

  1. Establishing key performance indicators are aligned to business needs
    1. Business problem to Analytics problem justification
    2. Definition and justification of the metrics used to evaluate the business problem
  2. Quantitative analysis identifying key business insights
    1. Analytical approach directly addresses the business need
    2. Description of how data was processed (e.g. cleaned, manipulated, transformed, etc.) to ensure analysis is broadly applicable.
    3. Performance evaluation of analytical approach(s) is provided.
  3. Providing unique insights and recommendations for business improvements
    1. Key drivers of outcomes are provided
    2. Clear statement of recommendations is provided
    3. Recommendations are placed within the broader context of the business need
  4. Professionalism and salesmanship of solution
    1. Presentation is visually engaging and professionally delivered
    2. Business problem, analytical approach, and results are well connected and clearly articulated
    3. Recommendations and implications to the company are well-defined and appropriate based on the context

Challenge Question

What is the future of autonomous mail/delivery services across North America through 2030? Demonstrate key factors that can be exercised to develop scenarios, plans and planning systems that can be customized and made accessible to operators as well as fleet planners and managers.

Data

Exclusive access to external data will be made available to teams who wish to incorporate feeds from data sources made available via A2Go’s partnership with Demyst. Entrants who desire to have access to Demyst data should email lanhamm@purdue.edu with their contact information and their info will then be shared with Demyst Support Team who will provide permissions to their data catalog for use in the Hackathon.

Tools & Software

Any open source tools may be used.

About the Co-Sponsors

Analytics2Go logo

Analytics2Go (A2Go) delivers Machine Learning for enterprise level companies and small-to-medium-sized businesses. The platform applies augmented analytics to uncover hidden, valuable business insights micro-apps that are narrowly scoped for ease in deployment and adoption. Each microapp is dedicated to improving particular business decisions (such as reducing customer churns, improving sales forecast accuracy, identifying operational problems, and diving into their root causes, etc.) through specifically designed augmented analytics. The A2Go microapp catalog is accessible to any size business or industry subscriber.

INFORMS Data Mining Section logo

The mission of the Data Mining Section is to promote and disseminate research and applications among professionals interested in theory, methodologies, and applications in data mining and knowledge discovery. We have over 500 members coming from industry and academia. We have held a highly attended workshop the day before the INFORMS Annual Meeting for the past 13 years, where the top scholars, practitioners, and up and coming talent present their work.

Questions

Contact Matthew Lanham at lanhamm@purdue.edu or Aaron Burciaga at aburciaga@analytics2go.com.