Predicting the success of construction waste removal flights
at the time of flight launch

AI-based systems

construction

5%

95%

the number of successful flights has increased

by

the accuracy of flight success forecasting has increased

solution

Technical solution

1. Conducted a study, identified patterns and factors that influence the success of waste removal

2. Machine learning (ML) and data analysis methods were applied:

Using the open-source TabNet and LGBM libraries, we implemented two flight success prediction models. The models were trained on the customer's historical flight data and are capable of making accurate flight success predictions at the time of flight commencement.

We compared the models and selected the one with the highest accuracy

Result

Business values

Effective waste management

proactive response to potentially incorrect flights

The working group's time spent analyzing incorrect flights after their completion has been reduced by 3 times

reducing costs and minimizing financial losses from incorrect waste removal trips

A 5% increase in successful flights due to proactive response from dispatchers after the model's implementation

Identification of key factors

Understanding which factors and conditions influence the success of flights - constants and variables: carrier, driver, departure time, landfill, type of waste being removed, weather conditions, permit specifications, etc.

Accuracy

Development of software
and Big Data solutions

Send a request and our specialists will contact you within 1 hour.

Choose a convenient method of communication