We are almost at the end of 2024 and I want to come back to one of my old experiences that demonstrates how you can experiment with technology and check if you can create a working solution that generates value. One of the great ways of validating technology is taking part in a hackathon. You have a well-defined problem (sometimes unfortunately not related to your business domain), motivation as very often it is done as a contest and limited time for creating a solution. Those elements help you to focus on the main challenge and just limit disruption to the minimum.
Every year Objectivity, organises the IdeaApp, a 2-day hackathon during which 10 teams compete to build the most relevant and innovative solution. The 2022 edition was special for me, it was the first and last I participated. Moreover, my team’s goal was quite ambitious — we decided to build a quantum computing solution.
The challenge was enormous. In only two days, we had to deliver a working application. We wanted to connect to a quantum computer directly so we could show results in a near-time way. The plan was to use real data as much as possible.
This was the high-level initial idea. At this point, we just needed to find a business challenge to solve.
The Business Challenge
When working with technology, I always try to use it with business scenarios that can be identified in your real life and generate value for you from the first day. For the hackathon, we decided to try to help our CEO Rob Helle find the best flight plan to visit all our offices in December. Our goal was to provide the possibility of meeting every employee of Objectivity before Christmas and allow them to chat with the CEO.
From the user’s perspective, our CEO got the option to decide whether he prioritises the reduction of flight time, cost or CO2 emission. He was able to define how important each of these parameters is to him and the organisation by simply moving the sliders. The application also allows for selecting the place and date of both the start and the end of the journey, as well as the minimal number of days to spend in each location.
Based on the input of the CEO’s preferences, people’s availability and flight data, the application will propose the optimal flight plan. The summary screen includes the percentage score in the aforementioned three key parameters, information on all flights, and people you‘ll be able to meet over the course of your journey.
In the example screen you can see above, the priority was to find the cheapest flight plan. As you can see, you can adjust the criteria to find flight plans that will be shorter and more eco-friendly.
The Technical Solution
From the technical perspective, we needed to select the technologies which would allow us to match the business requirements develop a working prototype within just the two days of the hackathon.
We started with the quantum part of the equation. We decided to use D-Wave as the provider of quantum computing power for our solution. Our team created a QUBO (Quadratic unconstrained binary optimisation) problem that described the constraints defined in this challenge. After testing and fine-tuning, we were able to use it directly on a quantum computer with the use of API calls. The result was calculated nearly in real-time, so we could integrate it with the application directly.
For building the classical part of our application, we selected the Streamlit framework. It allows for creating data apps rapidly, and we needed to be fast in order to complete the solution during the hackathon. I was amazed by how fast we made the app that looks nice and provides all the necessary features for gathering and presenting data. Also, we were able to use a GitHub repository to store the code and coordinate work within the team, as it automatically supports CI/CD pipelines. The application was being compiled and released just a few seconds after committing the new code.
The last challenge was to find flight data. This turned out to be more difficult than we initially anticipated. Our plan was to use an API that provides flight information, including flight plans, costs and CO2 emissions. Unfortunately, we didn’t find anything that could be used in the hackathon.
Most APIs that provided that information required passing an application process, and they clearly stated that they’re offering data primarily for commercial purposes. Our intended use of this data was even mentioned as prohibited in some cases. Because of that, we needed to find another way of getting the real data. So we created a scraper that extracts the data from the Google Flights page.
The last piece of the solution was the information on people’s availability. For the hackathon’s purpose, we decided to create the possibility of managing that in our application. Of course, the next step would be an integration with Objectivity’s HR system.
In total, these decisions allowed us to build a working application in just two days.
The Outcome
From the business perspective, the solution allows for selecting the flight plan that will support multidimensional organisation criteria. Previously, while planning a trip, we optimised only flight time and cost. This optimisation is often basic — selecting the cheapest or shortest flights. By using the quantum computing approach, we managed to build a more complex goal. It not only includes cost and flight time, but also the sustainability factor represented by the CO2 emission. With this solution, the organisation can define the weight of each component and get the flight recommendations in a matter of seconds.
Our solution is an excellent example of what can be achieved with the current quantum devices. Let’s try to generalise the described challenge. It includes the date, cost and ticket availability of the flight, travel duration and CO2 emission. Additionally, it takes into consideration the employees’ availability calendar. The CEO wants to meet with everybody in the manner that allows for meeting the defined goals. It’s an excellent example of a scheduling or planning challenge. Think about the potential of such a solution in planning activities in a factory, production plant, or other business areas.
From a retail perspective, you can think about optimising supply chain logistics. For manufacturing, you can improve efficiencies of your production lines. And in healthcare, you can try to optimise the shift scheduling and usage of expensive medical equipment to help as many patients as possible. These are just a few examples that can easily be extended to different industries or business cases.
Leave A Comment