12. Procedure Automation Lifecycle - Operation

During normal plant operation, operators rely on automated procedures to control processes and equipment. While these automated systems are running, engineering staff should continuously monitor how well the automation is performing (tracking metrics like response times, setpoint accuracy, and system stability). Regular assessment activities and operational audits must also be scheduled to evaluate overall system performance.

Purpose: These monitoring and audit activities serve two key functions: maintaining and improving the automated control procedures, and verifying that the systems continue to meet their original design specifications and performance requirements.

Note: Automated procedures in this context include everything from basic PID control loops to complex sequential control schemes and safety interlocks that run without direct operator intervention.

12.1 Operator Feedback

Standard Operating Procedures (SOPs) that have been reviewed, tested, and approved by operations teams still need ongoing refinement to fix errors or improve process methods. These improvement opportunities typically surface during two key phases: operator training sessions and actual production runs.

Definition: Operations input refers to feedback from floor operators, technicians, and production supervisors who execute the procedures daily.

Control systems should include an easily accessible feedback mechanism that operators can use during both training and live production. This feedback loop is critical because operators often identify practical issues that weren’t apparent during the initial procedure design phase.

Industrial Example: An operator might discover that a valve sequencing step in the procedure doesn’t account for the actual response time of a large control valve, requiring a longer delay between steps.

Note: The feedback collection method should be integrated into your HMI (Human Machine Interface) or accessible through simple documentation tools to encourage regular use without disrupting production workflow.

Improvement opportunities in control system procedures can vary from simple adjustments to significant overhauls. When collecting feedback on operational procedures, ensure you capture the following key information: Procedure Name/Identification, Section/Step Number, Problem Description, and Proposed Solution. This systematic approach helps track issues and implement solutions effectively across different control processes.

Note: Problem descriptions should be specific enough for other engineers to understand the issue without having witnessed it firsthand. Include details about system behavior, parameter values, and operating conditions when the problem occurred.

The following are examples of typical problem descriptions you might encounter:

Equipment Misidentification Error

The control system is configured to operate the wrong piece of equipment due to incorrect device identification. This typically occurs when device tags, I/O addresses, or communication parameters don’t match the actual field equipment.

Common Causes: Incorrect P&ID references, wrong device addresses in the PLC/DCS configuration, or mislabeled field instruments during commissioning.

Industrial Example: A PLC program intended to control Pump A (tagged P-101) accidentally operates Pump B (P-102) because the output coil address points to the wrong motor starter, potentially causing process upset or equipment damage.

Example: The wording in a prompt is ambiguous.

When system prompts or operator instructions contain unclear language, control systems may interpret commands incorrectly, leading to unexpected behavior or safety issues. Ambiguous wording can cause the system to select unintended control actions or fail to execute the desired response.

Industrial Example: A prompt stating “increase temperature when pressure is high” could be interpreted multiple ways - should the system increase temperature immediately upon detecting high pressure, or only maintain higher temperature while pressure remains elevated? The lack of specificity in timing and magnitude creates operational uncertainty.

Note: Clear, unambiguous language in control system prompts is essential for reliable automation and operator safety. Always specify exact conditions, timing, and magnitude when writing system instructions.

This procedure meets the required operational objective but has room for optimization to achieve better efficiency. The current implementation works as designed and delivers the expected output, however alternative approaches or parameter adjustments could reduce execution time, energy consumption, or resource utilization.

Note: When evaluating procedure efficiency, consider factors like response time, steady-state error, overshoot, and computational load on the control system.

Industrial Example: A temperature control loop that maintains setpoint within ±2°C but takes 15 minutes to reach steady state could be retuned to achieve the same accuracy in 8 minutes through improved PID parameters or feed-forward compensation.

The control procedure fails to meet specifications because it cannot achieve the required process setpoint or performance objective.

Note: This typically occurs when the control strategy has fundamental design flaws, such as insufficient actuator authority, poor sensor placement, or inappropriate controller tuning that prevents the system from reaching its target operating conditions.

Industrial Example: A temperature control loop that oscillates around the setpoint but never stabilizes within the required ±2°C tolerance band, making the procedure unsuitable for the application requirements.

Operations management reviews change requests and prioritizes them against other process control and automation modifications currently in the queue. The process control team then responds to approved requests by coordinating with Process Engineers or Run Plant Engineers from the Procedure Owner organization to gather necessary feedback and technical requirements.

Note: The Procedure Owner organization refers to the department or group responsible for maintaining and approving the specific process procedures being modified.

Implementation follows standard software modification protocols (such as development, testing, and validation phases) while maintaining strict configuration control conventions. Configuration control ensures that all changes are properly documented, tracked, and reversible, which is critical for maintaining system integrity and regulatory compliance in industrial environments.

Definition: Configuration control conventions are standardized practices for managing changes to control systems, including version control, change documentation, backup procedures, and approval workflows.

12.2 Monitoring

This activity reports on how well the automated control procedures are meeting their performance targets for the key metrics (measurement criteria used to evaluate system performance) that were defined in section 8.1.10 and Annex A.

Note: This essentially tracks whether your automated sequences are hitting their setpoints and staying within acceptable tolerances for parameters like response time, steady-state error, and overshoot.

Operators must understand both what an automated procedure is designed to accomplish and its current execution state (whether it’s running, paused, completed, or has encountered an error). A Technical Reviewer should be designated to evaluate and identify enhancements for the procedure itself or the human-machine interface (HMI) that operators use to monitor and interact with the procedure.

Note: The execution status typically includes real-time feedback on procedure progress, alarm conditions, and any manual intervention points where operator input is required.

This evaluation process should ideally occur in the live production environment where actual operating conditions and constraints are present. However, a high-fidelity training environment that accurately simulates production conditions can serve as an acceptable alternative when production testing isn’t feasible due to safety or operational concerns.

When evaluating a control system, the assessor should focus on these critical areas:

System Performance: Review how well the control loops meet their setpoint tracking requirements and disturbance rejection capabilities. This includes examining steady-state accuracy, transient response characteristics (rise time, settling time, overshoot), and overall system stability margins.

Safety Systems Integration: Verify that all Safety Instrumented Systems (SIS) and Emergency Shutdown (ESD) systems are properly configured and tested. Check that safety integrity levels (SIL ratings) meet the required risk reduction factors for each safety function.

Control Strategy Documentation: Ensure process and instrumentation diagrams (P&IDs), control narratives, and logic diagrams accurately reflect the implemented control strategy. This documentation should clearly show the relationship between field instruments, control systems, and final control elements.

Tuning Parameters: Assess whether PID controller parameters (proportional gain, integral time, derivative time) are appropriately tuned for each control loop. Look for evidence of systematic tuning methods rather than trial-and-error approaches.

Alarm Management: Review alarm rationalization studies and verify that alarm priorities, setpoints, and response procedures follow recognized standards like ISA-18.2. Check for alarm flooding scenarios and nuisance alarms that could overwhelm operators.

Human Machine Interface (HMI): Evaluate operator displays for clarity, consistency, and adherence to high-performance HMI principles. Screens should provide situational awareness without information overload.

Cybersecurity Measures: Confirm that network segmentation, access controls, and security patches are properly implemented according to standards like IEC 62443.

Note: Each evaluation point should include verification that the system meets both functional requirements and applicable industry standards.

• Does the operator understand what is happening during the performance of the procedure?
• Does the operator understand the meaning of all prompts requiring operator action?
• Does the operator act as directed by the procedure or does he/she feel compelled to perform a “workaround”?
• Could efficiency be increased by having the control system perform more of the tasks that the operator is performing (without sacrificing operator understanding)?
• The number of interactions and awareness notifications may make a procedure so tedious that it is inefficient, and the operator does not pay close enough attention to important process activities.

Another critical aspect of the assessment is performance measurement, which serves as the foundation for control system optimization. The evaluation should examine whether product quality can be enhanced through improvements to the automated control procedures currently in use.

Real-time monitoring must run continuously alongside automated operations. This can be implemented through operator dashboards or automated report generation systems that provide regular performance updates. Purpose: The goal is to detect process deviations and implement corrective actions much faster than traditional scheduled maintenance audits or periodic reviews.

Industrial Example: This continuous monitoring approach might reveal that gradual equipment wear (such as valve seat erosion or sensor drift) has altered key process parameters, requiring control loop retuning or equipment maintenance before product quality is significantly impacted.

Note: The monitoring system essentially acts as a supervisory layer that watches both the process and the control system performance, enabling proactive rather than reactive maintenance strategies.

A raw material change, such as a change in pellet size, may occur and affect a chemical reaction, but the impact on the automated procedure often goes unnoticed initially. By monitoring changes in procedure execution time (the time it takes for the automated sequence to complete), control engineers can identify when an automated procedure needs updating or retuning.

Why this works: When raw material properties change, reaction kinetics and mass transfer rates typically change as well, causing the process to deviate from its expected timeline. This timing deviation serves as an early indicator that the control strategy may need adjustment to maintain optimal performance.

When designing control systems, it’s important to consider various optimization approaches to improve overall performance. Different methods can be applied depending on your specific objectives - whether you’re trying to minimize settling time, reduce overshoot, improve steady-state accuracy, or enhance disturbance rejection.

Common optimization methods include: - PID tuning techniques (Ziegler-Nichols, Cohen-Coon, or model-based approaches) - Pole placement for desired transient response characteristics - LQR (Linear Quadratic Regulator) for optimal control with weighted performance criteria - Frequency domain shaping methods using lead/lag compensators

Note: The choice of optimization method should align with your system requirements and constraints. For instance, a temperature control system might prioritize minimal overshoot to prevent equipment damage, while a servo positioning system might focus on fast settling time and tracking accuracy.

12.3 Operational Audit

An operational audit is a lifecycle activity that should be conducted at regular intervals (as defined in your facility’s automation philosophy document) or when specific events trigger the need for review. The primary purpose is to verify that your automated control procedures are functioning as designed and to preserve the reliability of these automated systems.

Definition: An automation philosophy document outlines the standards, guidelines, and best practices for how automated systems should operate in your facility.

Having clearly defined performance metrics and consistent monitoring practices will make these audits more efficient and actionable. When you establish good measurement practices upfront, you can quickly identify issues and verify system performance during audit reviews.

Note: Effective metrics should be measurable, relevant to safety and production goals, and tracked consistently over time.

Typical metrics that control engineers should consider tracking include:

• Number of operator intervention
• Number of times the procedure automation runs
• Number of alarms triggered during the automation
• Quality of associated product
• Change in resource utilization

An operational audit can uncover gaps in procedure automation that aren’t obvious during normal plant operations. These gaps typically fall into two categories: instances where the current automation philosophy (the overall strategy governing which procedures should be automated and how) isn’t being properly implemented, or situations that highlight opportunities to enhance the automation philosophy itself.

Note: Procedure automation gaps often become visible only under audit conditions because day-to-day operators may develop workarounds or manual interventions that mask underlying automation deficiencies.

Industrial Example: An audit might reveal that operators routinely perform manual valve adjustments during startup sequences, indicating either a gap in the automated startup procedure or a need to revise the automation philosophy to include these critical manual steps.

The work processes and procedures that ensure compliance with the procedure automation philosophy (the overall strategy for how automated systems should operate and integrate with manual processes) should be evaluated for effectiveness periodically or when triggered by an event such as a system failure, safety incident, or major process change. The audit should review work practice documentation which may include:

Note: This evaluation helps identify gaps between your intended automation strategy and actual implementation, ensuring that automated procedures continue to meet operational and safety requirements as systems evolve.

• Verification that automated procedures take action to avoid defined consequences and achieve desired benefits.
• Documentation of automated procedure configuration.
• Documented procedure automation specification.
• MOC documentation of modifications to procedure automation configuration and required control system parameters.
• Procedure automation performance monitoring reports.
• Documentation of repairs to malfunctioning automated procedures.
• Documentation for out-of-service automated procedures.
• Consistency between automated procedures across similar units.

When audit processes identify control system problems or deficiencies, you need to create structured action plans to address each issue. These action plans serve as your roadmap for resolving problems systematically and ensuring nothing falls through the cracks.

For each identified problem, your action plan should include three critical components. First, establish realistic timelines that account for equipment availability, maintenance windows, and the complexity of the fix. Second, assign clear accountability by designating specific team members or departments responsible for implementing each corrective action (this prevents the “everyone’s responsibility is no one’s responsibility” problem). Third, define how and when you’ll review the results to verify that your corrective actions actually solved the original problem.

Note: Timeline considerations should include whether the fix requires a plant shutdown, ordering of replacement components, or coordination with operations personnel.

Industrial Example: If an audit identifies drift in a temperature transmitter, your action plan might assign the I&E technician to recalibrate it within two weeks during the next scheduled maintenance window, with the control engineer verifying proper operation through trending data review within one week after completion.

This structured approach ensures that audit findings translate into actual improvements rather than just documentation, and provides a clear trail for future reference and compliance purposes.

12.4 Requirement

Note

The procedure owner (the designated engineer or manager responsible for this specific control system procedure) shall ensure that corrective action plans are developed for any issues, deviations, or non-conformances identified during audit processes. These action plans should include specific steps to address root causes, assigned responsibilities, and target completion dates.

Note: Audit processes in control systems typically include safety system testing, calibration verification, documentation reviews, and compliance checks against industry standards like IEC 61511 or ISA-84.