Ai2020 ai and or final
Download as PPTX, PDF0 likes2,267 views
1. Operational research (OR) can contribute domain knowledge and methodology to developing useful artificial intelligence (AI) applications for business. OR expertise includes scheduling, facility planning, forecasting, and other traditional applications. 2. OR's "soft" problem structuring methods can help address complex, "messy" business problems suited to AI. These approaches consider stakeholder perspectives and seek progress through learning. 3. OR is exploring how to build ethical AI, ensuring applications are fair, transparent, accountable, and respect human and societal rights. The OR Society and British Computer Society could collaborate on these issues.
Ai2020 ai and or final
- 1. Professor of Business Analytics, University of New South Wales Professor of Business Analytics, Birkbeck University Emeritus Professor of Systems Thinking, University of Hull AI2020: 8 December 2020
- 2. Chollet, F., (2018). Deep Learning with R. Manning Publications. Artificial intelligence Deep learning Machine learning Classical programming Rules Data Answers Machine learning Data Answers Rules Deep learning Chui, M., and McCarthy, B., (n.d.). An Executive’s Guide to AI.
- 3. Operational Research (OR) “The discipline of applying advanced analytical methods to help make better decisions” (INFORMS and ORS) Data Science “Data science is a multidisciplinary approach to extracting actionable insights from the large and ever-increasing volumes of data collected and created by today’s organizations ... presenting the results to reveal patterns and enable stakeholders to draw informed conclusions.” (IBM) Business Analytics is concerned with “the extensive use of data, statistical and quantitative analysis, explanatory and predictive models, and fact- based management to drive decisions and actions” (Davenport and Harris, 2007) Hindle, G., Kunc, M., Mortensen, M., Oztekin, A., and Vidgen, R., (2020). Business Analytics: Defining the field and identifying a research agenda. European Journal of Operational Research, 281(3): 483-490.
- 6. HARD approaches SOFT approaches Problem definition Seen as straightforward, unitary Seen as problematic, pluralistic Model A representation of the real world A way of generating debate and insight about the real world The organisation Taken for granted Has to be negotiated Outcomes Product or recommendation Progress through learning https://www.lancaster.ac.uk/users/incism/back2.html “Over the last 40 years, methodologies have been developed to deal with “wicked problems” or “messes” that are beyond the reach of the traditional, mathematical modeling methods of operations research (O.R.). These methodologies are structured and rigorous but non- mathematical. Examples include Soft Systems Methodology (SSM), cognitive mapping/SODA and the Strategic Choice Approach (SCA). Collectively they are known as Soft O.R., Soft Systems or Problem Structuring Methods (PSMs), and they have been accepted as an important part of O.R. almost everywhere with one notable exception the United States.” John Mingers (2009)
- 7. 1. Domain knowledge: the hard OR discipline has a wealth of knowledge and experience in traditional OR applications in which AI can be embedded 2. Methdology for business: OR can bring a business focus to building AI applications in organisations – in particular, soft OR can help in in structuring and acting in messy problem situations 3. Ethical: the OR Society is exploring an ethical perspective on the role of AI in organisations and its impact on citizens and society
- 8. Scheduling: of aircrews and the fleet for airlines, of vehicles in supply chains, of orders in a factory and of operating theatres in a hospital. Facility planning: computer simulations of airports for the rapid and safe processing of travellers, improving appointments systems for medical practice. Forecasting: identifying possible future developments in telecommunications, deciding how much capacity is needed in a holiday business. Yield management: setting the prices of airline seats and hotel rooms to reflect changing demand and the risk of no shows. Credit scoring: deciding which customers offer the best prospects for credit. Marketing: evaluating the value of sale promotions, developing customer profiles and computing the life-time value of a customer. Defence and peace keeping: finding ways to deploy troops rapidly.
- 9. Vidgen, R., Kirshner, S., and Tan, F., (2019). Business Analytics: a management approach. Red Globe Press. https://businessanalyticsmanagement.com Vidgen, R., Shaw, S. and Grant, D. (2017). Management challenges in creating value from business analytics. European Journal of Operational Research. 261(2): 626-639.
- 11. 11
- 15. The Information Commissioners Office (ICO) defines automated decision-making:
- 16. Algorithmic pollution “Algorithmic pollution is a form of social pollution which denotes the unjustified, unfair, discriminatory, and other harmful effects of algorithmic decision-making for individuals, their families, groups of people, communities, organizations, sections of the population, and society at large.” Cecez-Kecmanovic, D., Marjanovic, O., and Vidgen, R., (2018). Algorithmic Pollution: Understanding and Responding to Negative Consequences of Algorithmic Decision-making. Proceedings of the IFIP 8.2 Working Conference, San Francisco, USA, December 11-12.
- 17. Vidgen, R., Hindle, G., and Randolph, I., (2020). Exploring the ethical implications of business analytics with a business ethics canvas. European Journal of Operational Research, 281(3): 491-501.
- 18. Citizens Informed Understandable Contestable Organisations Fair Transparent Accountable Safe Society and the environment Rights (human and non-human) Social justice Vidgen (2020). Ethical AI and algorithmic decision- making Framework. Work in progress with ethicalAI and the OR Society.
- 19. Summary: what OR can do for AI Bring a wealth of knowledge and experience of traditional OR applications in which AI development is embedded Provide methods to guide the development of AI applications that create business value Constitute an ethical framework and ethical practices for those working in the development, management, and governance of AI applications Opportunities for the Operational Research Society (ORS) and the British Computer Society (BCS) to work together?