This is the story of an oil well and the data it provided throughout its lifetime.
The well, hereafter referred to as "F-12", began in 2004 with preliminary reports. Many events happened through its life, until its abandonment in 2016.
Here is an example of how you can look at an entity's entire lifespan of data and derive insights from that data using WellLine.
We'll start with searching for the well in our tenant.
From the Home screen, click the "Equinor" tenant:
Scroll down in these suggestions and select
The Timeline view for the well appears. We can now see the entire life of this entity across all data currently in the Equinor tenant.
In this case, the well has 5263 events across ~12 years of time:
To start customizing our view and looking for insights, let's first narrow our view to events that happened in 2007. To narrow our view, we simply drag the gray box (the TimeSlice Selector) in the LifeLine to just cover the events displayed in this period of calendar year 2007:
When we do this, all areas beneath the LifeLine (the TimeSlice area) automatically update to show only events during this time period:
Now, anything we do in the TimeSlice area will only affect the period of time we just selected.
To see what type of events occurred during this time period, let's look at the Event Filter panel. This panel provides aggregation metrics and allows for quick filtering based on an event hierarchy.
For example, we can see that many events defined as "Drilling" occurred, while almost no "Completion" events occurred. This gives an immediate indication that the well was drilled during this period of time.
There were also 116 "Interruption" events that occurred. These could reveal important issues that occurred during drilling, and perhaps even some lessons learned.
Let's see what specific types of interruptions happened; to do so, click the arrow next to "Interruption (116)".
Some of these event types are helpful in determining what happened, but 80 events are listed as "Other", which is not very descriptive. Luckily, WellLine offers other methods of figuring out what these events are besides just reading through all of them.
To start our investigation into what these "Other" events are, let's turn off all events except "Other". To do so, click the slider next to the "Event Types" header to turn off all events, then click the checkbox next to "Other"
The rest of the TimeSlice is now filtered to just these 80 events. To get a better idea of when most of the interruptions occurred, we can zoom in using the Histogram.
There are a couple ways to zoom the Histogram:
With a mouse, use the scroll wheel
With a touchpad, use 2-finger scroll or the touch scrollbar
Adjusting the gray TimeSlice Selector bar (as we did previously)
Let's zoom in near where the most light blue appears in the Histogram. It appears that the area around July 22nd contains many of them:
The Histogram also has the ability to overlay time series data, if available. To add this data, we can click the "Measures" dropdown and select one or more measures.
Let's add the measured depth ("depthMD") and wellhead elevation ("wellheadElevation") to our Histogram:
This effectively overlays a days over depth chart plus provides an indication of when the wellhead was installed:
This provides useful context:
The wellhead was installed on July 15th
The hole depth progressed several days later
There were many "Other" interruptions during this time
We can also use the Histogram to orient ourselves in time. Clicking the Histogram around the period of July 22nd will move the EventStream to this same period:
We could now read through the events that occurred during this time period. For example, open the last "Other" event from 2007/07/21 by clicking the up-arrow icon:
This opens an event card, which displays many event details, including which entities are part of the event. Some of these entities come from structured data--the well name and the rig name, for example--while many others come from WellLine's entity extraction processing.
In addition, clicking an event in the EventStream adds that event to our current URL. In fact, almost all of the filtering we have done so far is being captured in the URL.
For example, our current URL is something similar to the following:
This allows us to quickly and easily share our current view with others, fostering communication and knowledge gathering.
Closing the event card brings us back to the Timeline view:
Instead of reading through dozens of written drilling comments, we can use the power of entities to see summaries, patterns, and even potential events of interest. To do so, we use ConnectionViewer:
The dropdown options in ConnectionViewer represent groupings of entities. The number next to each group is the number of unique entities that it contains. For example, in the above screenshot there were nine (9) types of Equipment Problem entities found in our current view.
Opening a dropdown reveals the entities that make up the group. Let's take a look at what Equipment Problems were found for the "Other" interruptions we are currently investigating:
The number next to each entity is the number of events, in our current view, in which it occurred. This provides useful information in our investigation:
There were many repairs undertaken
Leaks, including hydraulic leaks, had to be addressed
One specific problem involved one or more mud pumps
Now that we have a better general idea of the actual interruptions, we can investigate specific subsets by using entities as filters to our current view.
Let's take a look at the mud pump issue; to do so, click its + icon:
This adds the entity to the Entity Filters container. We can add as many filters as we'd like to the container, applying them by clicking the "UPDATE & APPLY" button.
For now, let's look at this single event by clicking "UPDATE & APPLY" now:
Our Timeline is now filtered to just the single event where "mud pump problem" was found. The "1" in the Entity Filters container is the number of entities we are filtering by.
We have now identified and isolated a specific Interruption event that occurred while drilling this well. This event can be easily shared with others, recorded in a report, or used as a starting point for a deeper investigation into mud pumps.
Let's move on to the longest period of the well's life: when it was producing oil and gas. To do so, let's clear any filters and move our TimeSlice Selector to the production portion of the LifeLine:
Our previous selections in the Event Filter panel are remembered, so let's reset this panel as well:
Looking at the entity types, we can quickly tell that this period of the well's life was focused on production: the 2972 "Production" events are far higher than any other type of event.
Looking at the Entity Filter and the Histogram, we can also see that the well had several periods of "Workover" events, in 2009, late-2010, and late-2014:
Since there is production data available we can overlay production metrics in the Histogram. To do so, open the "Measures" dropdown and choose "oilProduction", "gasProduction" and "waterProduction":
This overlays the three measures.
There is a lot going on in the Histogram now, so let's make the Histogram panel easier to read by hiding "Production" events, hiding the Event Types panel, and resizing the panel to make it larger:
These are several of the ways to customize the Timeline view; adjust the view to your liking.
We can see that the oil and gas production declines steadily after 2010. However, in late-2014 there is a noticeable change to the production metrics. To see what caused this, let's zoom in to this period of time:
We can now clearly see that this period of time had an effect on production: water drops significantly while oil and gas rise significantly. To understand why, we can look at other portions of the WellLine interface and get a quick idea of these 69 events.
To see the full picture, let's un-hide Event Filters and see what type of events occurred:
Expanding "Workover" shows us the work that was performed:
There are 9 "Perforate" events. Were these the cause of the production improvement? Let's look only at these "Perforate" events:
A quick glance at EventStream shows that there were several successful perforations on 2014/11/26. The event filters were helpful here and allowed us to quickly find useful information.
What if these events had been given the type of "Other", or something less-helpful? Another way we could have found this important information is through ConnectionViewer.
Show all "Workover" events, then open the "Equipment":
Here we can see the various pieces of equipment that were mentioned during these "Workover" events, including "perforation gun", which is connected to 3 events.
Just because a piece of equipment was mentioned does not mean it was used, however, so let's add "perforation gun" as a filter (by clicking its + sign, then "UPDATE & APPLY") and see it in context:
The resulting view shows several events where this piece of equipment was mentioned, including one in which the perforation guns were "activated":
Through different WellLine interface capabilities we have effectively arrived at the same conclusion: intervention efforts in 2014, including a perforation of the well, resulted in improved oil and gas returns and a reduction in water returns.