Orchestrating Advanced Process Control Programs
Doug McCallum
Value-Based Business Development | Advanced Process Control (APC) Evangelist | Cherished Mentor | Trusted Adviser | Builder of APC Implementation Programs | APC Benchmarking Authority
The process industries have significant optimization opportunities across all aspects of their mill, refinery, or plant (site). Typically, these projects deliver an economic payout within 3-6 months. Many companies struggle to manage and maintain enterprise-wide Advanced Process Control (APC) programs and lose out on these significant returns. This article introduces utilizing the “APC Score†KPI to develop and manage large-scale APC programs. APC Score is industry and technology-agnostic, size and complexity-independent, and an overall APC key performance indicator. APC Score strongly correlates to the benefits achieved; therefore, improving your APC Score will enhance your site’s profitability. Utilizing APC Score improvement for project prioritization will significantly improve your program's outcome.
APC benchmarking can help identify and quantify gaps in a site’s advanced process controller deployment and performance. With this information, you can develop an APC program to direct resources to close those gaps and capture additional value. APC benchmarking can also be utilized to track program progress and provide data to make better decisions that will improve your site’s profitability.
"You?can't?manage what you?can't?measure.†Peter Drucker
Utilize Benchmarking to Plan and Track the APC Program
The process industries comprise numerous site configurations; no two mills, chemical plants, or refineries are alike.? A method to benchmark these diverse sites needs to be industry and technology-agnostic, size and complexity-independent, and allow a site to track its performance versus other similar sites anonymously.? A benchmarking study should show where you are doing well and where potential improvements can be made.?
APC benchmarking comprises multiple KPIs to help identify key APC gaps, such as the percentage of Benefits Achieved, APC Score, Coverage Score, Scope Score, On-Stream Time, Controller Effectiveness, Technology Score, and Organization Score.?
% Benefits Achieved
Each process unit’s benefits are derived by the percent the controller is on while allowing all the manipulated variables (MVs) degrees of freedom to push the unit to its economic optimum. What impacts a unit’s controller effectiveness are operator training, controller technology, controller design, controller maintenance, instrumentation, and regulatory loop performance. In addition, the controller’s scope can impact the percentage of benefits achieved. To reach a unit’s true potential, the controller must be designed to be unit-wide to capture full benefits.
APC Score
APC Score is industry and technology agnostic, size and complexity independent, and an overall APC key performance indicator. The higher the APC Score, the higher the benefits achieved; therefore, improving your APC Score will directly enhance your site’s profitability.?
Breaking down the APC Score into its components helps identify where these benefits are being lost, whether from incomplete deployment across units, inadequate controller scope, reactive maintenance practices, or a lack of enabling technologies to improve engineering productivity.
Coverage Score
Incomplete deployments will show up in the Coverage Score.? This KPI is the % of units that could have APC and have APC deployed.? For example, a site has ten units on which APC can be applied. ?However, if the site has only deployed APC on four units, then that site’s Coverage Score would be 40.
To improve its coverage score, a site must embark on an APC adoption program to accelerate the benefits accrued through the implementation of APC.
Scope Score
The Scope Score is the average of all individual unit Scope Scores.? A unit’s Scope Score is the % of manipulated variables (MVs) a unit should have versus the controller’s current design.? This KPI checks to ensure the controllers are designed to be complete unit-wide controllers.? For example, a unit’s full APC potential scope is 30 MVs.? However, if the current controller design is only 15 MVs, then that unit’s Scope Score is 50%.? If the unit does not have APC, it is not included in the average Scope Score.
A low Scope Score can be improved through controller redesigns and revamps.
On-Stream Time
On-Stream Time is the percent of the time the controller is on when the unit is available.? The unit should be considered unavailable during these possible conditions:
- The unit or part of the unit is down or is in a startup or shutdown mode.
- Some of the process equipment is damaged and will be repaired in the short term.? If the plan is to operate for a long duration with damaged equipment, it should be considered available, and a retest should be performed.
Controller Effectiveness
Controller Effectiveness is an important KPI since it directly relates to the achieved benefits.? It should be an online KPI that will show how the controller is performing.? Controller Effectiveness is Controller Utilization multiplied by Controller Quality.
Controller Utilization
This KPI shows how unconstrained the controller is and allows it to reach the economic optimum.
- Use data only when the controller is on (ONSTS).
- Count the number of controlled variables (CVs) at high/low limit, setpoint, ramp, or external targets (CCS).
- Count the number of MVs at external target or engineering limits (MFU).
- Count the number of MVs at minimum movement, in wound up, in bad status, or are taken out of service by the engineer (MOK).
- The actual number of MVs in the controller (IPMIND).
- Unit Utilization = (CCS + MFU + MOK) /IPMIND * 100.
- The site’s Utilization = Average of the Unit Utilizations.
Controller Quality
This KPI quantifies how well the controller predicts the controlled variables (CV).? Choose an equal number of CVs as MVs, which are the most typical constraints.
- Use data only when the controller is on (ONSTS).
- Calculate each CV's baseline Standard Deviation (CVSD) with the controller off. CVSD could be calculated before the project commences or when the controller is off.??
- Calculate each CV's dynamic performance factor (DPF). DPF = 2*[CVSD—average of the absolute value of the (variable—its steady state target) ]/ CVSD.
- The Controller Quality KPI is the average of all its CV’s DPFs.
- The site’s Controller Quality is the average of all units’ Controller Quality with APC.
Controller Effectiveness
A unit’s Controller Effectiveness is its Controller Utilization multiplied by its Controller Quality.? The site’s Controller Effectiveness is the average of all the individual unit Controller Effectiveness with APC.
Causes of low Controller Effectiveness can be from:
- Bad controller design:? The original scope and/or tuning was not done adequately.
- Operational mode has changed:? The controller was designed for one mode, and the unit operates in another. The controller has not been updated, causing the operator to clamp the MVs.
- Poor APC technology: ?The core APC technology cannot handle the problem or lazily responds to disturbances, and operators clamp the MVs to keep the controller stable.
- Lack of maintenance: ?The process has changed, and the models don’t match the current operation, causing the operator to clamp the MVs to stabilize the controller.
- Operator acceptance:? Operators do not trust the controller; therefore, they clamp the MVs.
- Equipment issues: The site is operating in an unusual operating envelope. Depending on the length of time the issue is expected to exist, a retest might be necessary.
Controller design and maintenance significantly impact Controller Effectiveness. Utilizing advanced adaptive technologies helps increase Controller Effectiveness by enabling proactive maintenance programs. As the chart shows, technology score directly correlates with Controller Effectiveness.
Technology Score
This KPI is based on the quantity of advanced technology that has been deployed in the pursuit of capturing APC benefits:??
- Automatic testing
- Adaptive advanced process controller
- Inferential building tools
- Adaptive inferentials (virtual online analyzers)
- Performance monitoring data historian
- Virtual assistants
- Smart KPIs defined
- PID loop tuning tools
领英推è
Adopting advanced technology to empower your engineering staff can improve a low Technology Score.
How to Estimate Benefits
While not the focus of this paper, a basic understanding of a simplistic way to estimate benefits is needed.? The five main areas where APC improves a process unit’s operation are increased capacity, better yields/conversion, energy usage, catalyst consumption, and chemical use.? Depending on the unit and operating goals, these three can be used to estimate the APC’s value.? For example:
- Capacity Benefit = Capacity Increase % Unit Margin % Time Pushing the Unit * Capacity
- Yield Benefit = Yield Increase % Yield Improvement Margin Capacity
- Energy Benefit = Energy Improvement % Cost of Energy Total Energy
These benefit calculations help identify the value of the performance gaps.
Gap Analysis
Once the component scores have been determined and compared to the best-in-class performance of the top 3 similar-sized sites’ gaps, the value of closing those gaps is calculated.
Develop a Program to Close the Gaps
Now that the gaps have been identified and the value assigned by closing those gaps, what should you do?? Looking at each unit’s economics and metrics, we can financially prioritize the new projects, revamps, and migrations list.? Ideally, focus on the upper right-hand corner and work to the origin. Approaching your program in this matter will maximize its value for the site.
Program Plan
The leadership is onboarded during this phase, and the teams are selected. A training program for the project engineers is implemented, and organizational changes are made to support the program. The initial deployments of APC are defined, and staffing plans for both the technology supplier and client are finalized.
Initial Unit Deployment
Start multiple projects; several could be first of a kind for the customer and require a successful demonstration of APC on that application.? Develop the scale plan by incorporating lessons learned during this phase.? Identify, recruit, and train additional engineers for the Scale phase.
Scale
Deploy as fast as possible while developing a proactive APC maintenance practice for long-term APC sustainment.? Identify additional areas of opportunity to deploy and capture more value.
Program Governance
Proper program governance is the conductor of the implementation orchestra, communicating and coordinating multiple projects to ensure successful outcomes. Critical elements of governance are:
- Formalize Executive Sponsorship, members of the Steering Committee, and the frequency of meetings.
- Develop and align a formal project plan with other mill initiatives to ensure overall resource balance.
- Utilize structured, monthly project reports, maintaining engagement with mill leadership.
- Conduct quarterly program update sessions with senior corporate management (the Steering Committee).
- Design and implement formal program communication and marketing campaign to drive regular communication with crucial mill employees to update on progress, lessons learned, and benefits captured.
- Continuous monitoring through a standard APC KPI dashboard that reports online performance and dollar benefits obtained.
- Track APC Score progress with annual benchmarking.
?
People and Organization
One of the most significant success factors for an APC program is building an APC culture and deploying empowering APC technology.
The following points effectively segregate and describe this varying sophistication along the organizational capability spectrum:
- No Advanced Process Controller Organization: There is no formal group to implement and maintain advanced process controllers. The site might have a dedicated individual, who utilizes whatever advanced control technology was previously purchased. APC performance is not actively measured.
- Static Advanced Process Control Organization: This is a dedicated advanced process control team with a lead engineer. There is no career advancement once one enters the group. They have standardized on advanced control technology. They create rudimentary APC performance metrics and management reports.
- Robust Advanced Process Control Organization: The dynamic team comprises APC experts, young process engineers going through their rotation, and operations experts at the mill.? Seeks proven innovative control technologies to gain a competitive advantage.? For multi-site organizations, a corporate leader and group have been established.? The team utilizes sophisticated performance metrics, and management reports such as online benefit calculations
Improving organizational capabilities increases the success of advanced process control, maximizing potential benefits. There is a significant economic incentive to implement APC and keep it at peak performance.
Once a company has decided to invest in advanced process control, a supporting organization must be developed to ensure the program's success and the ongoing capture of benefits.
- Dedicated Team: This team consists of advanced process control experts, young process engineers on rotation, and tech-savvy operators. The number of manipulated variables per control engineer varies from 50 to 200, depending on the individual’s capabilities and the process's needs. Having a mixed team promotes skill development and knowledge transfer, helping to ensure success and keeping the controllers at maximum performance.
- Independent Budget:? The control group maintains a separate budget from the process units’ operating budget for new controllers and controller revamps.? An independent budget keeps controller projects from being deferred, allowing the complete capture of potential benefits.
- Communicate:? A tight-working relationship between the control group, operations, planning, and scheduling allows for communicating constraints, turnarounds, and economic targets amongst the groups.?
- Metrics on Performance: Besides the simplistic on/off %, comprehensive metrics are captured, such as controller utilization, controller effectiveness, key constraints, benefits being derived, etc. Such metrics facilitate identifying poorly performing models within a controller and visibility to refinery management.
- Adaptive Technology:? Utilize adaptive technology with the ability to test while optimizing and minimizing the impact on operations, allowing the controllers to be frequently updated.? Deploy easy and intuitive technology, enabling young process engineers to impact the bottom line quickly.
- Training Program: The company sends new control engineers (young process engineers) for additional training.? The individual is paired with an operator and has a designated expert mentor.? These activities help develop new control engineers, furthering the organization's capabilities.
Summary
Determining where you are versus your potential and peers can help you develop a program plan to close your APC gaps and improve your site’s profitability.? Many benefits can be achieved; let’s find out where you are on the spectrum. Merely contact McCallum Value Partners today to schedule a benchmark study.?
“All successful companies are constantly benchmarking their competition. They have to know what they have to match up with day-in and day-out if their company is going to be successful." — James Dunn
Doug McCallum
Principal and Founder McCallum Value Partners
?
Enabling organizations to scale Computer Vision !
7 个月Thanks Doug. Well articulated as always.
Principal Solution Consultant
7 个月Love this. Understanding and deploying an actionable APC benefits assessment is one way stake holders can gain confidence on their APC investment and ensure long term sustainment.
Head Customer Success @ Phi Tech Solutions (AspenTech) | Industry 4.0 & Digital Transformation Enthusiast | Aspen Certified | Chemical Engineer
7 个月Very helpful! This cleared up many of my thoughts.
Engineering Manager, Fertilizer Division
7 个月very nice article. Indeed DMC-3 has some of these KPI's inbuilt within the APC. As for the profit calculation, what I have seen is something like a total differential where the contributions are derivatives of cost functions respect the change in a MV. I think that unless you can have within the LP optimization strategy the load, the reduction in process variability will not convert to profits unless you put pressure on the panel operators, who by the way, appreciate that the process is less variable, and can rest upon that, instead of pushing the process towards the limits.
Consultant- APC at iOG | Ex - BPCL | MGMCET'19
7 个月Thanks Doug....your post is really informative and gives a exact insights of key aspects in terms of sustaining benefits from proper and improper APC implementation in any case.