Project Overview
Oil Production forecasting is one of the most important problems in the Oil and Gas industry. Accurately estimating the amount of liquid that could be produced by a certain well along its lifetime is very crucial especially from the economical and business side. There are several factors that contribute into it, starting from the well location, geological factors, drilling technologies used by the operator drilling, neighboring wells and their effect on each others.
All the wells we are dealing with are located in the United States. For this milestone together with the coming ones, we are going to work on estimating the oil production for a certain well, location or group of wells using the data that we going to supply. Problem will be modeled differently from milestone to another. Ending with the productization of the output models.
The typical behavior of an oil well production is shown in the figure below. The production rate keeps on increasing for the first few months, until it reaches a peak. Then, it starts to decline. Production months before the peak are conventionally ignored due to their usually unpredictable behavior. After the peak month, an oil well starts in the decline phase, which extends throughout  its remaining lifetime. Throughout this project, we will only consider the decline phase. Any statistics, visualizations, or forecasts will only deal with decline phase, and the pre-peak months will be completely removed in the pre-processing phase.
Milestone Description
In this milestone you are going to use Machine Learning to predict three years cumulative production for a certain well. Given the production curve for a well, you should be able to build the label and make use of given well features to build a ML model that estimates the three years cumulative oil production. Well features could be engineering features, geological features, and location features. Well features will be found separately on two files, Harmony files and IHS files.