Train a dataset related to oil& gas industry using lstm for finding the optimum number of test that should be conducted in a month for each oil wells

I have a simulated data set of the oil and gas industry. This data set contains data of 365 days for 36 wells. Flow rates of each well are noted down for 365 days. My aim is to find the optimum number of test that needs to be carried out on each well in a month with a minimum number of errors. Possibly by using LSTM. Screen shots of the dataset