Cut down the costs of maintenance like power bill using elevator maintenance smartly, outlier prediction of applicance usages, sprinkler usage smart plannning etc

What it does

We have built four different models to tackle each of the problems mentioned above. We have also calculated the savings that will be incurred from saving using these models to be the order of 1000's of $ approximately 250,000$

How we built it

We have simulated some data using data sources available online and some background knowledge of the appliances.

Challenges we ran into

Creating the data set is very challenging as we need to have the domain knowledge of each problem. Besides that the algorithms that are to be used for each problem are very different and challenging to come up with the optimal ways of tuning the model to cost good amount of money

Accomplishments that we're proud of

Most of our solutions are ready to deployed and we have worked out our feature engineering thoroughly using different aspects of knowledge and also using the various similar datasets that are available on the internet. We have a very good estimate of savings that the company can incur adopting to the model solutions we have suggested. It is also very easy to deploy to production as we have made it easier for people to understand and deploy.

What we learned

We have learnt a lot about how to handle the open ended problems, gathering the data optimally, fine tuning the feature selection, the best model selection, tuning the hyper parameters etc. It is a whole some project where we were challenged at every step of the process of data science model building life cycle.

What's next for Data Maniacs

We would like to continue working on this, if you find it interesting we can catch up more and even pursue internships doing such cool work where we can have such great impact on clients and as well to the company saving millions of dollars with every optimisations we implement successfully.

Built With

Share this project: