3.2 Performance Measurement & Optimization

Reliability and maintenance teams in all manufacturing plants have a common goal; that is, to plan and execute initiatives that will help sustain the inherent reliability of the physical assets and increase the availability of the plant. Productivity and profitability of the manufacturing plant and overall organization are highly dependent on the reliability and availability of the plant. A thorough understanding of the importance of reliability has made the top management of major corporations invest in reliability and physical asset management. When the top management invests in reliability, they typically set the corporate strategy and directions for reliability through a road map for all the manufacturing plants that operate under the corporation. When the commitment or support from the top management is available for reliability initiatives, the onus is now on the individual manufacturing plant to develop a reliability plan that aligns with the corporate strategy and/or reliability road map and the current needs of the plant. The reliability and maintenance teams typically build all the strategic, tactical, and operational reliability plans that have initiatives that would make the biggest impact on continuous improvement in reliability and bring the desired benefits for the site. This presentation will explain the systems thinking approach, along with the seven golden rules and seven key factors, that will help reliability and maintenance teams to build an effective reliability plan. In addition, this presentation will also address the top three challenges in building a reliability plan and how to overcome those challenges through two case studies. The first case study is about developing a three-year reliability plan, and the second case study is about developing an annual reliability plan. Both the case studies will explain the application of the systems thinking approach, seven golden rules, seven key factors, and top three challenges that were dealt with and solved.

Recent developments in AI, ML and related techniques see wide adoption in many industries. However, in the asset management area, such technical advances are still in their infancy, especially in the maintenance, repair, and operations (MRO) area. Part of the reason is that, contrary to production, MRO data has its unique characteristics (e.g. incompleteness, inconsistency and heterogeneous), and most organizations are still planning to introduce diagnostic sensors dedicated to maintenance and equipment reliability. We face both challenges and opportunities in advancing data-driven continuous improvements within the asset management world. This presentation shares the findings of our current research and development focus. Titled “How to use historical data to find opportunities to improve the effectiveness of equipment reliability programs, optimize MRO inventory operations, and enhance MRO workflow management processes”, we will first examine the characteristics of MRO data as their uniqueness to a specific company, plant or equipment and their commonality across all sectors. Then we evaluate the feasibility of applying AI/ML techniques with MRO history for better operational efficiencies. We need to understand what data is related to human knowledge, human interaction and process, and what data is associated with the actual condition of the asset, and if there are patterns and models that can be learned. Last, we will demonstrate that AI/ML can find equipment agnostic models and patterns which help continuously improve MRO operations across different industries. Based on the findings, we will also show how AI/ML models learned from historical MRO data can be translated into prescribed actions for improvements in equipment reliability, MRO inventory and workflow operations for individual organizations.

The energy landscape is shifting with the rise in electrification of transit and the rise of renewable energy shaping a new energy era that is changing the way we think about infrastructure decision making. This presentation will articulate how electrification of transit and an increase in renewables will impact medium and long-term infrastructure planning by providing examples and a practical perspective (case study) to demonstrate how Asset Management decision-making played a vital role in a utility company’s response to this change. This utility company is a key contributor to several electrification initiatives. They recognized the challenge associated with these initiatives and the overall success of the first implementation phase with minimal disruption to current operations. They are also preparing for electrification of the government transit’s first all-electric bus garage to support future procurements of battery-electric buses (eBuses) and will be working on the design and implementation of charging systems infrastructure across the city(?). Over the past 20 years, more than 50 renewable energy systems have been installed on City buildings and properties. In 2020, the city developed recommendations for the utility to achieve greater outcomes for energy efficiency, demand management, and renewable energy. The city also mandated installation of renewable energy systems on all buildings, where feasible, by 2020. The rate of development in electrification and technology in the transit sector is faster than implementation of major infrastructure developments; changes in demand patterns impact everything from the transmission and distribution networks to generation, dispatch and peak-load system capacity design; so it is not possible to “wait and see” before committing to infrastructure investment decisions. This presentation will cover how the utility is dealing with these changes by ensuring an appropriate long-term decision-making framework is in place to assure business continuity and reduce the impact on climate because it poses a particular risk for asset owners and operators. AMCL will present best practices for long-term decision-making and how the impact of change should be taken into account during the development of long-term infrastructure planning processes, in the context of a public utility.

Is your company struggling to forecast, and have parts arrive on time to perform preventive maintenance, or know when downtime should occur that would be least disruptive? Would your company benefit from more accurate cost predictions and a decrease in last minute emergency breakdowns? Or perhaps you cannot depend on what the system has for inventory or how many staff are required on site at a given time – there are somehow too many or not enough at any given time. These are common issues faced by many large manufacturers from Marine, to Mining to Oil & Gas, and many others. This session will provide an overview of how basic Machine Learning techniques can be applied to Supply Chain, Reliability, and Asset Management to gain increased insight, and better overall performance. But what is Machine Learning? It is a subset of Artificial Intelligence, where algorithms improve through experience (new data sets). The algorithms constantly evolve as more and more data is run through it. Machine Learning is useful for finding unknown patterns and relationships in data, such as sales, plants, store, or forecasting. It is an effective tool to gain insight and efficiency in day to day operations, while also providing a future forward view.

L'objectif de ce webinaire, sera un partage d'un récent mandat chez un client minier ou ils ont fait l'évaluation de la maturité de la gestion de maintenance. Le webinaire couvrira également les découvertes et nos actions pour l'améliorer.Agenda de la présentation:- Contexte opérationnel- Évaluation (analyse de données et temps d'outil)- Découvertes- Action en partenariat- Résultats préliminaires- Les actions prochaines pour fiabiliser les opérations

The Northwest Territories has 33 geographically dispersed communities, many of which are not grid connected for community power supply. The majority are powered by small, local diesel plants with others supplied by three hydro systems which generate the majority of the total power produced annually by the Northwest Territories Power Corporation (NTPC). This paper will describe how the condition of these assets is assessed and monitored, the process used and how the results form part of the annual capital planning and implementation process. Asset management at NTPC is divided into Thermal, Hydro and Transmission and Distribution; the focus of this paper will be primarily on thermal and hydro asset health inspections. On an annual basis, one third of the assets are inspected in person by the relevant asset manager and a small team of subject matter experts. The process by which these inspections are conducted will be outlined, including what information is gathered, and what we look for when carrying out the inspections. The asset condition information for various component parts is entered into one spreadsheet for each asset. The paper will describe how an asset health index for that asset is assigned. The condition of the asset determines what further action or inspections are required and the criteria used will be discussed. The presentation will also outline how the results from the inspections are used as part of a project prioritization process for capital planning. Current reporting practices on individual asset health indices and for each power plant will be described. In discussing this topic, NTPC sees an opportunity to receive feedback from others and to promote discussion on best practice from other power utility companies.

ARMS Reliability was engaged by a client to vet its design for an extension of its oil facility with special focus on the diluent recovery unit performance. The facility had capacity to load and ship approximately 100,000 barrels of oil per day and were looking to increase the capacity to 120,000 bbls/d with the addition of a DRU. In order to meet capacity goals, ARMS Reliability assisted in building a Reliability Block Diagram (RBD) remotely using failure data from previous RBDs equipment types with similar operating contexts and assigning failure models using RBDs from other sites and the ARMS Component Strategy Library. Since design choices were not finalized at the time of build, multiple scenarios were simulated using existing VRU packages from other sites and data on reliability performance of previously modeled equipment at other terminals.This case-study presentation will discuss how a Reliability Block Diagram helped our clients:• understand expected performance of current and potential design choices• enable cost-benefit-analysis to determine what changes need to be incorporated for top contributors• target optimized strategies against top contributors to availability and capacity losses• quantify the best-case impact of process cleaning activities to inform their cleaning intervals and condition monitoring methods

Not knowing what you do not know can be very dangerous for an organization. With unfortunate events that led to injuries at competitor’s facilities, Skeena Bioenergy activated a safety review of all equipment using bowtie analysis and a risk matrix. Bowtie analysis identifies causes and preventative action to stop a defined event from occurring. Then, looks at loss prevention actions to prevent disastrous consequences that stem for the described event. The risk matrix is a chart that has frequency of occurrence on the vertical plain and the consequence of severity on the horizontal plane. When combined, gives a risk level number, colour coded, that identifies levels of acceptable and unacceptable risk. This application was successful in identifying that the design of the Cooler, one part of the process, does prevent fires and explosions. Further fire control measures identified will be added; 1. to improve containment of a fire so it remains in the Cooler and 2. to prevent a fire event from cascading into an explosion. These continuous improvements in the Cooler reduce the risk level to 3, Skeena Bioenergy’s acceptable level of risk. This abstract demonstrates the application and findings of applying bowtie analysis and a risk matrix to a piece of equipment, the basis of good risk management.

Asset management processes are becoming more widely adopted across industrial facilities. In this presentation, we’ll examine an essential aspect of the overall process—asset health—to understand the value of using data-centric models and how asset health enables fact-based decision-making at the asset and asset class levels across the enterprise. As a part of Intelligent Operations—a new approach to achieve Operational Excellence—asset health today uses digital transformation to optimize production, minimize equipment downtime, enhance human performance, and manage operational risks. We’ll examine the key asset-related aspects of Intelligent Operations and explore an asset health strategy based on the principles of interoperability and real-time decision support. Outcomes supported include reduced maintenance costs, enhanced asset availability, changes to predictive repair and capital replacement strategies, improved production, and reduced risk.

New technologies regularly enter the world of asset management (AM), often leveraging new inventions, which, in turn, are driven or supported by other advances such as computing power and big data handling. Such is the case for Motion Amplification, a new technology impacting crucial aspects of vibration analysis, machinery and structural troubleshooting, root cause analysis, and communications. This presentation will outline which areas of AM are impacted so we can have a road map facilitating the integration of the new tool into a global strategy and get an overall picture of the impacts it will have. We’ll also provide a brief technical introduction and some practical illustrations.Originally presented at MainTrain 2020