A challenge for importers in the automobile industry is adjusting to rapidly changing market demands like personalized features and e-mobility. In a new paper titled „Gone in 30 Days! Predictions for Car Import Planning“, Know-Center and Porsche Co & AG describe the results of their joint COMET project.

 More precisely, it’s a practical study of car import planning based on the monthly car registrations within Austria is presented. The project partners model the task as a data-driven forecasting problem and implement four different prediction approaches.

One method utilizes a state-of-the-art deep learning model which was compared to a traditional statistical model, while the other is based on LSTM-RNN. Both have been evaluated against basic linear and seasonal baselines. In the experiments, Know-Center and Porsche evaluate 33 different brands by predicting the number of registrations for the next month and for the year to come. The study will be published in the coming weeks in the „Journal for Information Technology“ which focuses on the topic of decision support.

More information here