Page 205 - Intelligent Digital Oil And Gas Fields

160 Intelligent Digital Oil and Gas Fields

achieving these levels include process cost and lack of the right level of
expertise in IT to implement the DOF solutions which have hindered
digitation in the oil field with only 1% of production data reaching people
to make decisions (The Economist, 2017). Unfortunately, today as a whole,
the E&P industry is somewhere between levels 1 and 2, with some semi-
predictive capabilities (level 4) in certain operational areas, such as, gas-lift
optimization, ESP, plunger lift, and fracture operations, through the use
of full-physics models coupled in a closed-loop IAM optimizer.
On the basis of the degree of smart components (logical solutions with-
out human intervention) and the complexity of the process to integrate and
orchestrate data types (categorical, logical, numerical, integer, and string),
workflows can be classified as four different types and can be grouped as illus-
trated in Table 5.2 which shows that for each workflow level, its data fre-
quency, primary tasks, collaboration, and integration with other disciplines,
physical model associated with the task, and types of actions:
• manual (100% human intervention)
• semiautomated
• automated:
controlled by human operators and
autonomous
• smart workflows (up to 20% human intervention; enable decision):
self-controlled
cognitive and self-trained
Fig. 5.7 depicts a production plot versus time to show the benefit of DOF
implementation at different level of complexity of automated workflows.
The production profile, depicted with red line represents a manual or
semiautomated process (A) without any form of DOF implementation,
all the processes are offline, engineers make decisions to change operation
settings, the production is affected by excessive downtime and the response
time is longer than 24h. The second system in Fig. 5.7 is the automated but
nonoptimized system (B). The process is online and generates rapid diagnos-
tics. The engineers maintain the control, but the behavior of the system is
not sustained. The third system is the automated and optimized system (C).
The process is fully connected to IOT and coupled to many software appli-
cations to allow optimization in real time; however, the operational settings
are controlled and supervised by the engineers. The most sophisticated level
is the automated, optimized, and self-controlled (autonomous or supervised)
system (D), which generates sustainable production gains.