Introduction to Asset Management in the 21st Century
We will introduce some of the international standards commonly used in Asset Management (AM), such as ISO 55000 standards, GMAM documents, the AM anatomy developed by the Institute of Asset Management, the International Infrastructure Management Manual, and the latest products of the Asset Management Council of Australia. We will explain how they apply to you and your organization.
Then we will discuss how to define and set the right policies in asset management, including how to set SMART goals for your organization and for your specific assets and how to mark your progress towards those goals over time to achieve world class performance.
Finally, we turn to questions of leadership and cultural change. How can we manage all this new technology and its potential applications, and what will it mean for our people? We will still need people to perform tasks and managers to define performance requirements. How will this work in our field, as AM advances into an increasingly technological world?
Background to Asset Management (AM)
- Why we need AM
- What AM intends to achieve
- International standards, documents, and frameworks, including ISO 55000, GFMAM, AM framework, IAM AM anatomy, IIMM, etc.
- AM program structure and components
Asset Management Policy and Strategy
- How policy and strategy work together
- Defining organizational goals
- Transforming goals into action through strategy
Asset Management Objectives
- Objectives vs. goals
- Using the AM strategy as a basis for long term implementation and sustainment
Asset Management Plans
- What are the various life cycle AM processes, and are they all relevant to you?
- Defining how to manage the various life cycle AM processes
- Applying processes and implementing strategy in each asset class using technologies and various tactical approaches
Leadership and Cultural Change
- What technology and its applications mean for the workers in an organization
- How to manage change, including organizational culture
Performance based contracts
- The significance of performance-based contracts
Basic Concepts in PAM
We will review the foundational concepts that enable the use of maintenance and condition monitoring data to make optimal asset management decisions, potentially saving companies millions of dollars. We will explain the use of probability distributions (and the Weibull distribution in particular) as powerful tools to describe and predict asset health over time. We will also offer some detailed procedures for using limited data to make optimal replacement decisions.
Another dimension of asset management is inspecting the asset or collecting condition monitoring data and using those readings to detect pending expensive failures and make appropriate actions to manage them proactively. For protective devices, it is necessary to periodically inspect them to ensure there are no hidden failures, and they will function in the case of an emergency to prevent costly consequences of multiple failures. With assets equipped with sensors or those with regular condition monitoring measurements, the data can be used to provide information on the health of the asset; this, in turn, is a critical tool for capital replacement planning or fit for service analysis.
Basic Concepts of PAM
→ Analysis of component failure data
→ Component replacement procedures
→ Reliability improvement through inspection
→ Life cycle costing management
Basic Conceptions in Machine Learning
The course will cover some of the most fundamental machine learning methods. C-MORE has actively applied machine learning methods to interesting real-world problems, such as the categorization of power generation units according to reliability characteristics and anomaly detection in linear assets to optimize required maintenance actions. Specific topics include types of machine learning three perspectives, steps in machine learning project, foundations of machine learning covering probability, optimization and information theory. Finally, the class will discuss how to evaluate the performance of machine learning model, pitfalls and some of remedies.
Introduction to Machine Learning
- Computing and Big Data
- Data science, AI, Machine learning and deep learning
- History of AI and Big Data
- ML in Practice
Taxonomy of Machine Learning
- Main Theories
- Underlying Models
Steps in Machine Learning
- Data acquisition and preprocessing
- Algorithm selection
- Training and Evaluation
Optimization, Probability and Information Theory
- Random variables and probability distributions
- Common probability distributions
- Baye’s rule
- Gradient-based optimization
- Information theory
- Error measures
- Bias-variance trade-off
- Over-fitting and under-fitting