Procedure Automation for Continuous Process Operations ⌂ Table of Contents
Annex A

Metric Examples (Informative)

50% Summary Edition
Important Note

The metrics and targets below are ONLY EXAMPLES. Each organization or plant will determine its own metrics and targets depending on process type, operator skill, HMI capabilities, operating environment, and types of automated procedures. Customize metrics to your situation.

The 7 Example Metrics

No.DescriptionTargetWhy It Matters
1Percentage of times operators manually perform a procedure instead of using an available automated procedure0%Ensures business justification for the procedure is met — benefits only realized if automation is actually used
2Number of operator control interactions performed in a fixed period (average and peak)Organization-defined average and peak limitsManages operator workload — too many interactions increases error risk and fatigue
3Number of times an operator had to intervene (deviate, override, abort, change a step) when an automated procedure was run0Identifies problems with procedure design, inadequate training, or equipment problems
4Number of times support from engineering is required during execution per period0Operators should be self-sufficient — engineering support needed indicates procedure or training issues
5Number of times an automated procedure experienced an abnormal end of execution per period0Quality indicator — every execution should complete normally; abnormal ends indicate bugs or design flaws
6Average, maximum, minimum, and standard deviation of duration of entire procedure, tasks, or stepsProcedure dependent (establish baseline first)Identifies optimization opportunities; high variability indicates inconsistent execution
7Product quality and yield measuresProduct and procedure dependentIdentifies improvement opportunities, shift differentials, training needs, and equipment problems

Key Notes on Metrics Implementation

On Operator Control Interactions (Metric 2): Two targets are required — average and peak — because either alone can be misleading. A low average with a high peak (100 interactions in one hour) represents operator overload that the average masks. Both average and peak must be monitored. Operator interviews provide context that quantitative data cannot.

On Operator Interventions (Metric 3): Root cause analysis should determine which of three causes applies: (1) problems with the automated procedure itself, (2) improper operator training, or (3) instrumentation or process equipment problems. Each cause requires a different corrective action.

On Duration Statistics (Metric 6): Establish a baseline by measuring initial performance, then set improvement targets. High standard deviation indicates the need for standardization. Investigate unusually long maximum durations for occasional problems.

Annex B — Model Drawing Philosophy

Drawing conventions are based on Entity-Relationship Diagram (ERD) format. Model types are visually distinct:

State-Based Control step/transition notation: Steps = solid rectangles; Transitions = lines that may be Automatic (solid), Manual (dotted, requires operator), or Automatic/Manual (solid line with "M"). Transitions can be Normal, Forward (skip ahead), or Backward (loop back).

Sequence-Based Control allows multiple steps active simultaneously — shown with parallel branches that synchronize at a completion point.

Annex C — Front-End Loading (FEL)

FEL is the process for conceptual development of automation projects before major capital investment. Three progressive stages:

StageQuestion AnsweredCost Estimate AccuracyDeliverable
FEL 1 — Options StudyWhat are my options to achieve project goals?±40–50%Options analysis, recommended approach, feasibility assessment
FEL 2 — Feasibility StudyIs it feasible?±25%Conceptual engineering package, risk assessment, environmental assessment
FEL 3 — FEED / Basic EngineeringHow exactly will we build it?±10%Complete P&IDs, equipment specs, control system design, construction-ready drawings

FEL 3 is where procedure automation specifications (Chapter 9) fit in the project lifecycle. Good FEL leads to successful project execution. Poor FEL leads to cost overruns, schedule delays, and operational problems.

This completes the lifecycle: Strategy (Ch. 8) → Planning & Definition (Ch. 9) → Specifications & Design (Ch. 10) → Implementation (Ch. 11) → Operation (Ch. 12) → Retirement (Ch. 13).