Closed-Loop Execution vs Open-Loop Advisory: Integrating Geosteering into a Drilling Automation System

You are sitting in an offshore Aberdeen remote operations center watching a 1,500-meter horizontal North Sea lateral drift toward the lower reservoir boundary while LWD gamma ray logs stream via WITSML at 0.5 Hz. With UKCS operators aiming for zero-NPT well construction, manual steering changes introduce execution delays of 15 to 45 minutes per slide, reducing effective reservoir contact across thin target zones. Maintaining optimal placement within high-value sand packages requires evaluating how real-time geological models communicate with surface rig equipment. Moving from traditional advisory workflows to direct execution frameworks transforms geosteering from an observational discipline into a dynamic control input for modern rig instrumentation.

Architectural Mechanics of Advisory versus Closed-Loop Integration

Open-loop advisory architectures separate subsurface model interpretation from physical control of the drilling rig. In an open-loop configuration, real-time telemetry data, such as gamma ray, resistivity, and directional surveys, streams from downhole tools through mud pulse or wired drill pipe telemetry to a surface decoder. The data is transmitted over WITSML protocols to subsurface geosteering software located in a remote operations centre or on-site logging cabin. Geosteerers interpret structural dip, evaluate bed boundaries, and generate trajectory recommendations. These targets are communicated to the directional driller via text messages, automated software prompts, or verbal radio calls. The directional driller manually reviews the proposed targets, evaluates rig mechanical limits, and enters modified weight-on-bit, top drive speed, or toolface orientation parameters into the rig auto-driller console. This multi-step human loop introduces significant operational latency, often delaying command execution across several drilled meters.

Figure 1. Data pathways connecting downhole telemetry to surface rig control systems.

Closed-loop integration routes interpreted structural dips directly to the drilling automation system via OPC UA interfaces following ISA 95 Level 3 control standards, as highlighted in technical literature like Implementation of Drilling Systems Automation. In this architecture, geosteering software acts as a real-time boundary condition generator within an automated control loop. Rather than relying on human translation of spatial targets into surface setpoints, structural model updates are converted mathematically into dynamic vector corrections. The automation engine ingests these vectors, calculates necessary surface parameter adjustments using physics-based drillstring models, and sends control commands directly to the drawworks, top drive, and mud pumps over high-speed industrial buses.

Field deployments documented in A Case Study of Integrated Closed-Loop Automation show closed-loop execution reduces weight-to-weight connection cycle time variability across operational crews by more than 10%. By bypassing manual data entry and human decision latencies, closed-loop systems execute continuous trajectory corrections that keep the bottomhole assembly aligned with the target window.

The impact of execution latency on wellbore position can be quantified directly. Consider a horizontal section drilled at a rate of penetration $v_{ROP} = 40.0\text{ m/hr}$ where structural dip changes create a target deviation angle $\Delta \alpha = 3.0^\circ$. If the open-loop manual decision cycle introduces a delay $t_{delay} = 0.5\text{ hr}$ (30 minutes) before the directional driller executes a corrective steer command, the uncorrected true vertical depth deviation $\Delta TVD$ is computed as:

$$\Delta TVD = v_{ROP} \times t_{delay} \times \sin(\Delta \alpha)$$

Substituting the operational parameters yields:

$$\Delta TVD = 40.0\text{ m/hr} \times 0.5\text{ hr} \times \sin(3.0^\circ) = 20.0 \times 0.05234 = 1.047\text{ metres}$$

In a reservoir sandstone unit with total vertical thickness of 3.0 metres, a 1.047-meter vertical displacement due to decision latency consumes more than one-third of the available target tolerance. Closed-loop parameter execution eliminates this human delay, maintaining trajectory alignment within tighter vertical bounds.

Technical Evaluation Criteria for Rig System Integration

Integrating subsurface geological software with rig execution systems requires evaluating critical performance metrics defined across industry operational standards. Technical committees under the International Association of Drilling Contractors and SPE Drilling Systems Automation Technical Section (DSATS SPE Drilling Systems Automation Technical Section) have defined criteria governing automated control stability, network latency, and structural data interchange.

Latency bounds must be maintained below 2.0 seconds for high-frequency dynamic control loops as defined by the IADC D-WIS interoperability framework. Surface control loops managing weight-on-bit, differential pressure, and top drive torque operate on millisecond cycles. If geosteering model inputs update at low frequencies or experience transmission delays exceeding 2.0 seconds, control algorithm outputs become desynchronised from actual downhole conditions. High network latency causes control destabilisation, leading to parameter overshoots, toolface hunting, and mechanical stick-slip. Closed-loop integrations utilize optimized data serialization over OPC UA networks to guarantee that interpreted dip variations update downhole steer vectors within sub-second thresholds.

Borehole tortuosity risk increases when continuous dogleg severity computation is omitted from dynamic auto-driller command pipelines. When steering commands are issued in discrete, delayed steps, the control system applies aggressive toolface offsets to return the wellbore to target TVD. These abrupt directional adjustments create micro-doglegs along the lateral path, increasing mechanical drag during subsequent trip operations and reducing overall casing running efficiency. Continuous monitoring of actual wellbore curvature prevents excessive localized bending. At GeoMaster, we find that manual recalculation of trajectory bending between survey stations creates latency, whereas continuous dogleg severity computation provides immediate feedback to the rig control system to prevent excessive borehole tortuosity.

System reliability requires strict formal verification and validation protocols per IADC guidelines (SwRI_B10_2018_0102_VnV-of-Sensors-and-Systems-in-Drilling) to prevent unstable control loop hunting across formation boundaries. When crossing hard-to-soft lithological transitions, surface weight-on-bit control loops can become unstable if geological interpretation algorithms interpret mechanical rate of penetration spikes as structural dip changes. Robust verification protocols ensure that control algorithms incorporate physical boundary limits, torque thresholds, and signal filtering before transmitting automated steering commands directly to rig actuators. Early industry studies documented in Drilling Systems Automation - A Technology That is at ... emphasize that automated control loops must include deterministic fallback modes to maintain operational safety when telemetry signals drop.

Direct Architectural Comparison and Decision Framework

Choosing between open-loop advisory workflows and direct closed-loop auto-driller integration depends on structural formation risk, sensor telemetry bandwidth, and rig control network maturity. The functional differences across core engineering parameters dictate where each architecture delivers maximum field value.

Parameter	Open-Loop Advisory Architecture	Closed-Loop Execution Architecture
Control Loop Latency	15 to 45 minutes (human review, manual entry)	Sub-second to 2.0 seconds (direct OPC UA pipeline)
Operator Intervention Requirement	Continuous (directional driller executes all setpoints)	Exception-based (automation executes within safe bounds)
Trajectory Precision & Tortuosity	Moderate precision; risk of localized micro-doglegs	High precision; smoothed continuous dogleg severity control
Suitability for Subsurface Risk	Optimal for complex, faulted, or uncalibrated zones	Optimal for predictable thin-bed horizontal sweeps

Open-loop architectures win in high-risk fault-crossing scenarios where manual interpretation by subsurface teams is required before trajectory modification. When drilling through complex structural fault blocks, sub-seismic faulting can displace target zones unexpectedly, causing sudden changes in formation dip and lithology. In these environments, automated closed-loop algorithms may misinterpret formation boundary indicators, leading to inappropriate steering adjustments. Open-loop workflows provide essential human oversight, allowing geologists and directional drillers to synthesize multi-disciplinary data, verify structural correlations, and manually approve trajectory adjustments before changing rig control setpoints.

Closed-loop control wins in continuous thin-bed horizontal reservoir sweeps, delivering up to 17.3% ROP improvement in lateral sections according to industry field trial results detailed in Closed-loop drilling optimization system: implementation and field results from large-scale deployment. In homogeneous or structurally predictable lateral sections, human decision delays represent the primary constraint on drilling performance and target zone exposure. By directly coupling real-time geosteering calculations to surface drilling automation systems, closed-loop execution maintains the bottomhole assembly within narrow true vertical depth windows without slowing drilling progress. Operational results from automated offshore campaigns documented in Case Study: Drilling Automation Delivers More Gas in Less Time at Fénix Project Offshore Argentina confirm that automated parameter execution increases overall rate of penetration while preserving wellbore quality. Technical background on system architectures supporting these deployments can be found in Drilling-Systems Automation: Current State, Initiatives, and ....

To determine the appropriate operational architecture, subsurface and drilling engineering teams can apply a clear decision rule based on structural uncertainty and execution constraints:

If the target reservoir zone exhibits high structural uncertainty, frequent faulting, or uncalibrated log responses, deploy an open-loop advisory architecture to retain human geological validation prior to parameter entry. If the reservoir section consists of a continuous, well-mapped thin-bed lateral with real-time telemetry latency below 2.0 seconds, implement closed-loop integration with the drilling automation system to eliminate manual execution delays and maximize net pay exposure.