Take a Deep Dive into Specialized
Content through HFES Workshops
All workshops take place on Monday, October 4
8:30 a.m. - Noon
A Practical Introduction to the Functional Resonance Analysis Method (FRAM) in Healthcare
Mark Sujan, University of Oxford
Peter McCulloch, University of Oxford
In this workshop we will introduce participants to the Functional Resonance Analysis Method (FRAM) and the Resilience Engineering principles that underpin it. FRAM has been used in healthcare for about ten years, and is increasingly being recognised and accepted as an important HF/E method in the patient safety and healthcare quality improvement tool kits.
Participants will learn about: the difference between work-as-imagined and work-as-done; safety as the ability to succeed; the role of performance variability and dynamic trade-offs in everyday clinical work; practical application of FRAM in healthcare settings; and the FRAM Model Visualiser (FMV) software tool.
The workshop is aimed at HF/E practitioners and people interested in healthcare quality and patient safety, who might have heard of FRAM or are curious about it, but who do not as yet have had the opportunity to learn about the details of the method or the underlying principles. The practical demonstration of FRAM in the workshop will make use of the FMV software tool, but the exercises can be done both with and without the FMV software. FMV is freely available and works across different platforms.
8:30 a.m. - Noon
Designing Human-Centered AI Systems: A Cognitive Engineering Approach
B.L. William Wong, Middlesex University
Sam Hepenstal, DSTL
AI- and other typically black-box algorithm-based technologies are increasingly being applied to replace or complement the human in tasks that traditionally required human judgment in high risk, high consequence environments. It becomes increasingly important that users and those affected by the technology are able to challenge and ascertain whether the outcomes from such systems are valid and sensible. How can designs for such systems be crafted to enable humans to understand, challenge and ascertain the outcomes from such systems?
In this workshop participants will learn how we might address this question. By lecture and case-study examples, participants will learn how we might design for transparency, how we make important relationships visible, open to inspection and verifiable, and how we might include relevant elements of context into such designs. We will draw upon methods and principles from cognitive engineering (Norman, 1986) and human factors (e.g. Wickens et al, 2013) and show how they can be applied to the effective design of visual representations of AI- or algorithm-based systems to produce human-centred AI, or HCAI, systems.
1:30 p.m. - 5:00 p.m.
Journey Towards High Reliability – Tiered Huddle System and Use of Human Factors and Systems Engineering Methods and Tools
Lukasz Mazur, University of North Carolina
To prevent patient harm events healthcare organizations must have a robust infrastructure to learn from unsafe conditions, near misses, and medical errors. Using a case study methodology, we will review UNC’s innovative approach to create an organization-wide Tiered Huddle System (THS) supported by Human Factors and System Engineering (HFSE) methods to promote high-reliability, situational awareness, and value creation. The session will be divided into three parts. First, a step-by-step process to the development and implementation of the THS including but not limited to the strategy, implementation efforts, escalation process, and project selection and complexation approach. Second, we will demonstrate how we identified implementation drivers and barriers and implemented mitigation strategies. Third, we will demonstrate the integration between THS with HFSE methods and tools to further gain improvements in patient safety and patient safety culture. Learning objectives are:
Learn how to identify the drivers and barriers to THS’s daily issue escalation and develop the interventions that will be most effective in promoting/removing those drivers/barriers.
Learn how to run a daily THS and integrate HFSE methods and tools during improvement projects.
Learn to design an improvement project to increase daily issues, identify the role of staff (front-line staff, managers, directors, risk management, patient safety officer, quality, physicians, and senior leadership) in the design and implementation efforts.
1:30 p.m. - 5:00 p.m.
Shanae Chapman, Nerdy Diva
Apply Agile Design Thinking principles to create change and support equity. Brainstorm how to create more diverse and inclusive products, services, and workplaces. Demystify Anti-Racism in tech by exploring research and stories specific to the tech industry and startup ecosystem. Brainstorm and prioritize actionable outcomes for Anti-Racism in your organization that you can start crafting immediately. Take action with an Anti-Racism MVP to make the greatest impact on your employees, colleagues, and customers starting now. Continue the journey and plan for long-term success with backlog management and agile ceremonies to track progress.
9:00 a.m. - 4:30 p.m.
Behavioral Data Analytics with R
John Lee, University of Wisconsin – Madison
Linda Boyle, University of Washington
Anthony McDonald, Texas A&M University
Data analytics, machine learning, and the increasing demand for experts in quantitative user experience present challenges and opportunities for behavioral scientists and human factors engineers. Data analytics and machine learning draw on techniques that are unfamiliar to many behavioral scientists, but data scientists may be unfamiliar with many important features of behavioral data. This workshop provides practical skills in behavioral data analytics and also addresses important issues specific to behavioral data. Participants will learn data manipulation and visualization techniques. They will apply these techniques to exploratory data analysis, machine learning, and model understanding. The workshop includes exercises and examples using the statistical package “R” that include: complex data reduction, creation of machine learning models, selection of cross-validation techniques suited to behavioral data, and visualization of predictions, and techniques to make models understandable. The workshop also includes a survey of machine learning techniques, such as text analysis, and resources in R.