DETOUR

GIF
Example airport status (Argentina) from ICAO
Example of country restrictions (Greece) from Sherpa
Example of country restrictions (Australia) from Amadeus
**Detour** - Data transformation and Workflow
Technologies used to create **Detour**

Inspiration

Is it safe to travel again? Where can I go to have the best experience with so many restrictions in place? What travel deals are available? Those are the most common concerns that travelers have these days, preventing them from booking travel. The coronavirus has disrupted our lives, global economy, and put the travel industry at high risk for bankruptcies and job loss.

Inspired by the idea to restore traveler confidence and help the industry with global recovery, the PlaneJanes team developed the solution that will help:

a) guide travelers to find the destinations with the best deals that match their travel aspirations and make confident purchasing decisions
b) provide modern technology and data for airlines and their sales channel partners (GDSs, OTAs, TMCs etc.) to stimulate demand and generate revenue   

What it does

For the travelers Detour is an app that guides travelers like you to your next destination with confidence. The app collects the information you need about health and hygiene measures, travel experience, social factors, and environmental sustainability and presents it in an easy-to-understand format. Detour uses built in intelligence to help you decide your next vacation destination by recommending destinations that best match your travel aspirations, preferences, and compatibility—so you can discover the wonders that await when you arrive.

For the industry Detour  provides the most relevant airline pricing and retailing content directly from the source and integrated with the latest safety, airport, traveler experience, and sustainability data, delivered via API, NDC, or other traditional distribution methods. Built by airline industry experts with scalability in mind, Detour utilizes NDC, NGS, and pricing standards to speed up the implementation, industry-wide adoption, and time to market.

How I built it

Our team started out with a blue-sky discussion to identify what our ideal fare search would look like during the COVID-19 emergency and beyond. We identified what was important to us as travelers while determining where to go, when to go, and how to get there. We wanted a single place to answer those questions so we did not spend hours researching a destination only to find out we cannot go for one reason or another. Instead of requiring the passenger to input a destination, Detour would recommend destinations based upon user preferences. We then vetted our idea with family and friends to challenge and reinforce our hypothesis before we settled on a direction.

Once we determined our Detour, we researched the APIs made available to us via the hackathon sponsors to ensure we had the technology and data available to us to create our ideal product. Once the creative process was completed, we focused on the technical details of our app. We used Angular, MongoDB, and AWS technologies to aggregate the data from the various API’s that we used to develop our app. We used the following API’s:

• ATPCO - Fares, Universal Product Attributes (UPA), Reassurance Universal Product Attributes

• Amadeus - Destination information

• Sherpa - Procedure information by country

• SIATA - Restriction by country

• ICAO airport status - Restriction by airport

• airport-data.com - Codes, latitude, and longitude

• Airport API - Openings and health and hygiene status

We identified and subsequently created six new Reassurance UPAs that we needed to complete our hack. We created the following Reassurance UPAs:

• Social Distancing

• Increased Airport Sanitization

• Touchless Kiosks

• Facemask Required in Airport

• Temperature Check

We then worked to process and analyze the information gathered from the many API’s that we needed to use to successfully complete our hack. The API’s helped us to determine:

• If the destination is open

• What health and hygiene measures are taken by airlines, city, state, airport, and country

• If there is a compatibility match to the criteria selected by the user

When determining the compatibility match, we had to determine if the cities, dates, and destination type match to the passenger’s input request. We then assigned a score of “Good”, “Better”, or “Best” to the match of the preferences and concerns of the passenger.

Our design approach was to use wireframes to identify features and scope. We then implemented using stylesheets from the initial prototype design to ensure the look and feel that would entice passengers to travel. We used Angular 8.0 with Material Design to create our UI. We used JAVA code for data processing and JSON for the data aggregation.

We utilized machine learning to determine the fare recommendation compatibility for the passenger’s request. We created efficient algorithms to provide concise data in the simplest form from the vast amount of data that we collected from the multiple APIs that we called.

We used blockchain for collecting large amounts of valid data from multiple sources.

Now that we had a large amount of text. We had to understand the text that we were receiving. We needed to learn, interpret, and process text from various governmental bodies/entities to determine the status of a location and any restrictions that have been put in place by the local authorities.

Finally, we needed to ensure the accuracy of the data and keep the data up to date.

Challenges I ran into

We came across three primary challenges in our development of Detour: Data aggregation, creating an unbiased algorithm, and consolidating the data in a meaningful way to the passengers using our app. In addition, we knew that with this information, maintainability was key.

The first challenge was determining which data was available that was in a form we could utilize, along with the level of reliability and accuracy that we knew we needed. Each passenger concern requires at least one data source, so we needed to aggregate the data from the multiple sources to provide a concise answer to address each concern of the passenger. For each data element, we had to identify the source and verify the validity of the data source.

The success of our app relies upon passenger’s trust, so it is of utmost importance that our data is accurate. Maintaining the data is important especially when we receive conflicting information. We need to be able to determine what data is correct and provide the most current data for our users.

Trust is the foundation of the app, and we wanted to ensure that we provided an unbiased algorithm. Our scores are in relation to the passenger’s needs and preferences and do not necessarily reflect the quality of the services provided by any of the airport, airlines, or destinations.

Our final challenge was how to consolidate so much information into an app available to passengers via an online tool or a mobile app. We chose to leverage the ATPCO Routehappy Rich Content Reassurance APIs while expanding to include new UPAs to provide an easy and efficient way to communicate information to the traveler.

Accomplishments that I'm proud of

One of the goals of the Travel Scrum hackathon is to develop ideas that will reshape the future of travel in new and innovative ways. We asked our coworkers, family, and friends what they needed to know to be inspired to purchase travel. Every response was based in trust, they wanted to know:

• The airport and airplane are following recommended sanitation standards

• Their money is not going to be wasted

• They will be able to enjoy a healthy and enjoyable trip

Trust is an integral aspect of getting travelers back to traveling. The PlaneJanes took the importance of trust to heart in our drive to create a prototype of a product that we would want to use for ourselves and our loved ones when we are ready to purchase our next vacation.

Aggregating the massive amount of data was a challenge that we are proud and happy to have overcome. Providing this data in a concise manner that passengers will easily understand was an important and valuable accomplishment to the completion of our app.

What I learned

We learned the importance of thinking through and validating the idea. When we researched, we found that all similar apps required the passenger to provide the destination. Our product is unique in the fact that we provide recommendations based upon passenger preferences and machine learning.

Most of the members of the team have participated in other hackathons but this is the first time we have a presentation and design specialist on our team. Our designer was able to step into the user interface and user experience role very easily by using her presentation and design skills to create an interface that that is beautiful, functional, and intuitive. We quickly realized the benefits of having a designer on our team.

We learned the amount of public data available to consume is massive. Ensuring the data is valid and current is more challenging than we had expected going into this.

What's next for DETOUR

Detour will continue to inspire travel and guide you to new destinations. While there are many travel apps in the market with common features that aim to find you the best deals, they only work well if you know where and when you want to travel and if those destinations are open for tourism.

Not only does Detour take into consideration all the COVID-related restrictions, but it will continue to help you find the best destinations based on your aspirations which may change in the future as COVID restrictions increase or decrease.

The app can be enhanced with other data sets to include weather, crime score, low fare alerts; changes to health, airport, and government restrictions; “Save For Later” and comparison features; hotel, and rentals information and much more.