One of the most common ways engineers use PetroAI is to screen and evaluate M&A opportunities. These projects usually move fast - from the first look to an executive presentation in just a few weeks. Unlike an asset team that spends years getting to know one area in detail, a BD team has to be ready to evaluate all kinds of deals, whether they’re bolt-ons, non-op positions, or larger strategic packages across multiple basins. That means teams don’t always have a deep understanding of the area they are looking at.
In this example, I’ll walk through a BD deal evaluation in the Midland Basin, using PetroAI to get from raw data to an economic comparison in a way that’s both quick and consistent.
The Deal
The opportunity here is two sections in Howard County with existing PDP wells both on and offsetting the acreage. That makes a traditional type curve approach tough, since any forecast has to account for the offset production. Just how far offset we place the wells and how many wells we think we can put here will also affect our type curve.
The acreage is in a core part of the basin, so the big questions are: how many wells can I still place economically, and is there any real upside? To answer that, I’m going to use PetroAI’s machine learning model that’s trained on subsurface data across the Midland basin. Because it accounts for well interactions, it’s a better way to handle the situation.
Many teams using PetroAI will choose to enrich the model with internal data like tops or internal geologic interpretations, but for this run I’ll stick with the basin model as-is.
Setting It Up
First, I load the shapefiles for the sections into PetroAI. With that done, I can quickly explore production trends in the surrounding area. This gives me a sense of how wells are spaced, what kind of completions are being used, and how those wells have performed over time.
Next, I start laying out potential well sticks. I try a few different spacing assumptions, then adjust the layouts to make sure I’m accounting for the existing wells. The goal is to build a handful of scenarios that can be compared later. The gun barrel image below shows my baseline case – a simple, WCA and WCB development.
I can also set development timing here. If I want to model a staggered development plan, I can put in different completion dates, bring wells on in waves, or leave benches for later. Once I’ve got my scenarios ready, I launch the build. PetroAI then generates time series predictions with diagnostics for every stick in each scenario.
Reviewing the Results
With the scenarios built, I move into the inventory analysis dashboard. This is where I can see how the different options compare. Right away, I see that DSU X outperforms DSU Y across the board. I can also see how production builds over time as wells are brought online, which makes it easy to understand the development pace.
The final step is to layer in economics. I can see how the wells perform financially and get a PV10 for comparison. This makes it clear which layouts create the most value and which ones don’t hold up as well.
Wrapping Up
In just a short amount of time, I’ve gone from raw acreage with existing wells to a set of scenarios that account for spacing, timing, and economics. All the assumptions, sticks, models, forecasts, and economics are version controlled and saved in a build. This makes it easy to share with my team or come back later and review. For a BD team moving quickly on a potential deal, that’s exactly the kind of workflow I need - fast, consistent, and grounded in the data.
