0
$\begingroup$

Consider a dataset of thousands of car repairs that have been performed. In simplest of terms, the columns to consider are the time of year when it was broken (seasonal changes in demand for car repairs), type of damage to car (some damages take longer to repair than others), and the type of car (some cars are more difficult to work on).

I am inquiring as to the best fit for trying to model data of this format where you are predicting the repair time based off of timeseries and categorical data as the inputs. Keep in mind that the data does not have a constant period.

Example Column Names:

datetime | Type of Damage to Car | Type of Car | Repair Time

Any suggestions?

$\endgroup$
0
$\begingroup$

So the question is about to model the next repair date given the previous repairs.

If you have customer-specific data, where you have logged customer repairs to a specific customer, then it would be good idea to do time series, provided you have enough instances of customer repeatedly coming back. If you have this scenario, then you can use something like an RNN (Recurrent Neural Network [https://towardsdatascience.com/understanding-rnn-and-lstm-f7cdf6dfc14e]) or LSTM, whereby you feed in the date time, type of damage at every time step to then get a single output date time.

If you have not logged a customer ID to the repairs to identify which repairs have been carried out for particular customers, then you could easily use a standard neural network for this. Here you would train the model on all input data with the goal of getting a date time which equates to when the next repair is likely to happen.

In terms of data representation, you could represent the following features as the following:

To make the input for a given entry, we then concatenate these features into one massive vector.

Good luck!

| improve this answer | |
$\endgroup$
  • $\begingroup$ Thank you for the feedback, I think I may have mis-worded the question. What I am attempting to predict is the "time to repair". So basically when a new vehicle comes in on a given date, I have a row of data with "datetime | Type of Damage to Car | Type of Car" and I need to predict time to repair that vehicle. I've been trying a RNN with the timedata and hot-encoded vectors for categorical but the mean absolute error is higher than I'd like. Any additional considerations? (I did the same thing you recommended with the time data) $\endgroup$ – kylepapili Jun 24 at 13:08
  • $\begingroup$ Right, understood, so the task would then only vary in that you are now predicting the time taken (in minutes, for example) than predicting the next repair date. The above still applies. The high error could be explained by a multitude of things from the data to the error function used to train the model. To diagnose the issue you would need to edit your post to include the model architecture include loss function used to train the model and how generalisation performance improves over epochs (e.g. a plot of training and validation set accuracy, for example, over epochs). $\endgroup$ – shepan6 Jun 24 at 14:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.