Inspiration
We were attracted to NTTData's proposed challenge, because of the possibility to optimize a complex system like a hospital's supply chain and reduce carbon footprints.
What it does
It analyzes data from previous years to predict the needs and prices for the products for the next year and provide a centralized purchase plan so that it prevents stockouts and oversupplying, while making more efficient the transportation of the deliveries.
How we built it
We came up with a formula to predict the data for the following years, depending on a factor which adjusts to fit best the previous data found using numerical methods. Then, we worked with the database provided to predict new data based on that, and plot into graphs to facilitate its analysis.
Challenges we ran into
It was not easy to come up with the formula, we had to modify it several times until we found one which we could not improve anymore and thought represented well the predictions. Additionally, some of the data did not follow any rational distribution (like a hospital making a single purchase in 9 years), which made the analysis more difficult.
Accomplishments that we're proud of
The good prediction we could make for the following year which fits the data, thanks to the prediction formula. Additionally, we are proud of the ideas we had to approach the problems and provide solutions to it.
What we learned
We learned the difficulty of analysing data to predict the future, and some ideas to tackle it.
What's next for NTTDATA-J.A.R.S-Purchase plan for healthcare supply chain
Keep improving by enhancing the prediction method used or some other factors.
Log in or sign up for Devpost to join the conversation.