Past Webinars

Decision Support System for Product (2-wheeler) Allocation to Multi-Plants across India through Analytical Mathematical Model

Speakers: 

Dr. S. Devarajan, Sr. Vice President (Advanced Mfg. Engineering)
K Ravi, Associate Vice President (Manufacturing Engineering)
MA Raju, General Manager (Manufacturing Engineering) TVS Motors Company Ltd, Hosur, India. 

     

TVS Motor Company is a two and three-wheeler manufacturer of both ICE & EV vehicle, leading with global ranking of 4, in terms of market share, championing progress through Sustainable Mobility with state-of-the-art manufacturing facilities in Hosur(Tamil Nadu), Mysuru(Karnataka) and Nalagarh(Himachal Pradesh) in India and in Karawang Jakarta. TVS Motor Company endeavours to deliver the most superior customer experience across 80+ countries. We manufacture the largest range of two-wheelers, starting from mopeds, to scooters, commuter motorcycles, to racing inspired bikes like the TVS Apache series and the TVS Apache RR310 covering with appx 290 variants of 2-wheelers with an annual production capacity over 5 Mn/ann.

TVSM production planning department faces a challenge of production allocation of product & variants across 3 plants in India to maximise the contribution while demand is highly fluctuating. OR Community of Practice (CoP), a unique approach at TVSM since 2020, entrusted to resolute this challenge, in consideration to various parameters for maximizing the contribution. It is extremely challenging to manually work out the optimized solution approach that balances the profitability, market share and resource utilization while the demand fluctuates month on month encountering seasonal spikes. Hence, OR CoP decided to develop a Decision Support System using OR technique to “Maximize the contribution as a objective function” with identified decision variables and constraints. The results are encouraging with increase of 0.7% in delta profitability  for one of the premium scooter and subsequently expanded to use DSS across the entire supply chain including suppliers and dealers.

Watch Decision Support System for Product (2-wheeler) Allocation to Multi-Plants across India through Analytical Mathematical Model

    Ideal Train Profile – a Seat Capacity Allocation Tool for the 
    Reserved Passenger Segment on  Indian Railways

    Speaker: Dr. R. Gopalakrishnan, General Manager (Data Analytics), 
    Centre for Railway Information Systems, Indian Railways

          

    Over 23 million daily passengers, including 2.1 million in the reserved segment, travel in the 12000 trains operated by Indian Railways (IR).  IR faces a recurring challenge on meeting the demand for reserved tickets within the existing capacity of a train. A typical travel demand may fall in the one of nC2 origin-destination pairs on a train with n halts.  Many passengers are issued waitlisted (overbooking) tickets at the time of purchase emanating from mismatch in supply and demand for tickets for multiple Origin-Destination combinations on a train. 
     
    The revenue managers were hitherto basing their seat-allocation decisions based on a time-consuming analysis of historical demand data, and experience.  The Ideal Train Profile (ITP) utility is a Decision Support System (DSS) developed by the Centre for Railway Information Systems for assisting the revenue managers in deciding the optimal seat capacity considering multiple factors such as importance of the station on the train route, and preference for short/long distance passengers or being fair to all the observed origin-destination demand combinations. ITP has been implemented in over 2000 trains on IR, covering the reserved segment in all the inter-regional trains. The results are encouraging with substantial reduction in issue of waitlisted tickets. 

    Watch Ideal Train Profile – a Seat Capacity Allocation Tool for the Reserved Passenger Segment on Indian Railways

            

      Exploring, Understanding, and Deploying Emerging Technologies in Retail: A Portfolio Management Approach

      Speaker: Andrew Cron, SVP, Chief Scientist, 84.51° 

          

      It’s no secret that emerging technologies offer many benefits to help streamline business efficiency and boost revenue, but realizing those benefits can be risky and inconsistent.
      In this session, Andrew Cron, SVP, Chief Scientist, 84.51°, will address how to bridge the gap between potential transformation through emerging technology and strategic product impact through why intentional innovation matters.

      He’ll cover:
      1. The importance of focusing on the business problem and strategic priorities that need to be solved and built.
      2. The role of innovation teams.
      3. Innovation and risk management through diversifying portfolios across multiple dimensions
      4. How to improve the odds and spot opportunities to innovate

      Watch Exploring, Understanding, and Deploying Emerging Technologies in Retail: A Portfolio Management Approach

             

        Efficient Distribution of Compressed Natural Gas: A case of Sabarmati Gas Limited, Gujarat, India

        Speaker: Ramesh Kumar, Assistant Professor, NITIE

            

        Rapid industrial development and urbanization in Gujarat has led to increased number of vehicles and demand for compressed natural gas (CNG). This study develops a simulation model for efficient distribution of CNG from mother stations to daughter booster (DB) stations to reduce the dry out (no supply) time and increase the efficiency of the CNG distribution network. The simulation model analyses transportation and resource allocation scenarios of the CNG distribution network, such as pooling of Light and Heavy Commercial Vehicles among DB stations, adding more LCVs/HCVs, increasing storage capacity at DB stations, replacing Light commercial Vehicles (LCV) with Heavy Commercial Vehicles (HCV), and adding new DB stations. Implementing this model has an estimated savings of about INR 1.3 million per year for SGL.

        Keywords: Gas distribution network, resource allocation, simulation, fuel transportation, dry out time.

        Watch Efficient Distribution of Compressed Natural Gas: A case of Sabarmati Gas Limited, Gujarat, India

          The Journey from Credit Scores to the FICO Platform: How the Way We Build Software Today Will Change How You Work Tomorrow

          Speaker: Jeff Day and Dr. Oliver Bastert

              

          Fair, Isaac and Company was founded in 1956 as a credit scoring service. Today, FICO is a leading global provider of analytics software.  Key offerings include point solutions—many in financial services—as well as general tools for Artificial Intelligence, Machine Learning, Optimization, and Decision Rules. The current and future growth strategy is the FICO Platform, an integrated decisioning ecosystem for authoring, validating, and operationalizing business analytics. We will discuss the ongoing evolution from a service-based financial company into a platform-first software organization. We will also explain how adopting a platform-first software development culture today will affect the way analytical models are created and deployed in the future.

          Watch The Journey from Credit Scores to the FICO Platform: How the Way We Build Software Today Will Change How You Work Tomorrow

            Addressing Pay Inequity: The Building of an Analytics Startup

            Speaker: Dr. Margret Vilborg Bjarnadottir

                

            Dr. Bjarnadottir founded PayAnalytics which has grown from a research idea to a solution with users in over 40 countries on all continents, impacting hundreds of thousands of people. The organization now employs 20 people and closed early A funding late last year. In this presentation and fireside talk, Dr. Bjarnadottir will introduce her company, talk about the hurdles, the mistakes, the surprises and the successes.

              Watch Addressing Pay Inequity: The Building of an Analytics Startup

              Optimizing Population Management in Correctional Systems

              Speaker: Mohammad Shahabsafa, Tamás Terlaky and Anshul Sharma

              The Optamo team, in collaboration with Pennsylvania Department of Corrections (PADOC), have developed the Inmate Assignment Decision Support System (IADSS) to optimize the assignment of the inmates to facilities and schedule their rehabilitative programs. Implementation of IADSS has equipped the PADOC to efficiently and accurately make inmate assignment decisions based on legal restrictions and best practice constraints, including security, rehabilitation programing, medical and mental health needs, education and vocational requirements, and distance of the inmates to home. IADSS has enabled the PADOC to reduce the operational costs by $2.9 Million in the first year of implementation and the projection of the savings over five years is over $19 Million.

              IADSS has also been implemented at the Georgia Department of Corrections (GDC) and has been used for the daily assignment of inmates to facilities and housing units since July 2021. IADSS has enabled the GDC to save millions of dollars annually.

              To the best of our knowledge this is the first time that operations research methodologies have been incorporated directly into the business operations of correctional systems. The successful implementation of IADSS in correctional systems opens a rich and untouched area for the application of operations research and optimization methodologies.


                Watch Optimizing Population Management in Correctional Systems

                Towards Healing Healthcare – Value-Driven, Innovative and Agile Analytics

                Speaker: Janine (Coelho) Kamath

                The healthcare landscape across the world is on a transformational journey which is likely to accelerate recognizing the significant lessons from the COVID -19 pandemic. Many healthcare organizations are rethinking their business model and value differentiation. We can expect analytics, digital technology, consumerism and value-driven care to play a greater role in the delivery of care especially with the growth of telemedicine, hospital-at-home initiatives, and remote patient monitoring devices. Mayo Clinic has a rich legacy and culture of using analytics, operations research, technology and management engineering for sustained excellence in clinical care, research, education and operations. This session will highlight the strategic priorities, novel application, learning journey and key success factors related to leveraging analytics to deliver patient focused healthcare.


                  Watch Towards Healing Healthcare – Value-Driven, Innovative and Agile Analytics

                  Operations Optimization Applications at General Electric

                  Speaker: Srinivas Bollapragada 

                  During my career spanning over two decades at GE Research Center, I had the opportunity to develop solutions to several complex operations optimization problems facing the company. I will share some those solutions which we had implemented in a number of industries including aviation, renewables, energy, healthcare, rail, finance, and broadcasting.


                    Watch Operations Optimization Applications at General Electric

                    The Princeton 20: A Framework for Identifying and Managing AI Project Risk

                    Speaker: Steve Sashihara

                    For 40 years, Princeton Consultants has had the privilege to work with many of the world's most successful companies to improve their core business processes with AI and Optimization. Our clients have spanned virtually every industry: Transportation and Logistics, Life Sciences and Healthcare, Manufacturing and Distribution, Media and Technology, E-commerce and Retail, Education and Government, and Financial Services. Throughout this time, we have come to believe that the hardest part of any project is "Deployment" -- the phase where a promising innovation moves from the laboratory to broad field production. In this talk, Steve will present "The Princeton 20," an internally developed framework we use to identify and manage the most important risk factors for the Deployment of an AI/Optimization system, independent of industry or application.


                      Watch The Princeton 20: A Framework for Identifying and Managing AI Project Risk

                      Webinar slides

                      Building a Vibrant Enterprise Analytics Organization at Meta (Facebook)

                      Speakers: Dr. JP Loose and Dr. Julie Ward Drew, Meta

                      Meta's Enterprise Products team develops and maintains scalable products that power the enterprise. JP and Julie will describe the Enterprise Products Analytics organization, the roles within the team, how we recruit and retain talent, and how we enable step-level improvements in enterprise productivity.


                        Watch Building a Vibrant Enterprise Analtyics Organization at Meta (Facebook)


                        Using AI to Co-develop New Product Formulations and the Future of Flavor Innovation

                        Speaker: Robin Lougee, The National Academies

                        Each year, more than 30,000 new consumer products are launched. An estimated 80% fail. In this presentation, I will introduce “computational creativity,” present challenges in new product creation and share results from a novel AI system that helps product developers more efficiently and effectively create new product experiences.


                          Watch Using AI to Co-develop New Product Formulations and the Future of Flavor Innovation Recording


                          Advanced Analytics in the Land of MOORE’S LAW

                          Speaker: Karl Kempf, Intel

                          The Decision Engineering Group was founded at Intel Corporation in August 1987. For the first 20 years, we focused on manufacturing and supply chain problems, then in 2007  expanded our scope to include product design and development problems. I will briefly describe the major projects over this time. Then I will present a few of the basics of how the group is constituted – how we pick projects – our development process – interactions with our internal customers – and more. 


                            Watch Advanced Analytics in the Land of MOORE'S LAW

                            O.R. for Social Good Practical Implementations

                            Speaker: Dr. Natalia Summerville, SAS Institute

                            Natalia will provide overview of four Data4Good Optimization applications:

                            • Medical Resource Optimization for Cleveland Clinic
                            • Student Monitor Optimization for Boston Public Schools
                            • Collection Site Optimization for Mothers Milk Bank
                            • Back-to-School Optimization for Durham Public Schools

                              Watch O.R. for Social Good Practical Implementations recording


                              Driving Business Value at Ford through Analytics

                              Speaker: Erica Klampfl, Ford Motor Company

                              Ford Motor Company and the automotive industry have a long history of applying analytics to solve tough business problems. New technology, data sources and analytics methodology are unlocking business value and enabling Ford to address some of the most challenging problems facing the industry.  Erica will share her lessons learned from leading an analytics organization at Ford.

                              Watch the Driving Business Value at Ford Analytics recording


                              Deep Transformation with Smart Supply Chain Digitization

                              Speaker: David Simchi-Levi, MIT

                              The presentation is focused on a practical supply chain digitization strategy that includes three important components: a unified, single, view of demand; supply chain segmentation; and smart planning and execution, all of which are powered by Digitization, Analytics and Automation.

                              Watch the Deep Transformation with Smart Supply Chain Digitization webinar recording


                              Veritec Solutions Revolutionizes Self Storage Pricing Practices

                              Speaker: Dr. Warren H. Lieberman, Veritec

                              Using customer segmentation techniques in a new and innovative manner, we highlight an automated and dynamic capability known as Value Pricing. Without using purchase fences/restrictions, Value Pricing allows offering customers multiple rates for a product previously offered at a single price. By introducing the concept of “dynamic relative desirability,” Value Pricing enables greater revenue gains than the static inventory allocation approach used in some industries to differentiate inventory, such as the seat selection process for airlines, where certain types of seats are sold for premiums. We will show how Value Pricing is incorporated into the websites of several companies.

                              The increased revenues made possible by these methods is a “game changer” for many self-storage companies. The President of one company has described it as the “next step in self storage evolving and growing.”

                              Watch the Veritec Solutions Revolutionizes Self Storage Pricing Practices webinar recording


                              Building and Sustaining Analytics at Chevron

                              Speaker: Margery Connor, Chevron

                              Chevron and the energy industry have a long history of applying analytics to solve tough business problems. See examples in this article. With the business challenges and the technologies to solve them growing dramatically over the past decade, building an analytics team of the appropriate size and skillset has been a moving target. Margery will share her lessons learned from leading the analytics organization at Chevron.

                              Watch the Building and Sustaining Analytics at Chevron webinar recording


                              Predictive Models for Optimizing Imaging Decisions for Detection of Metastatic Prostate Cancer

                              Speaker: Dr. Brian Denton, University of Michigan

                              We used data-analytics approaches to develop, calibrate, and validate predictive models, to help urologists in a large state-wide collaborative make prostate cancer staging decisions on the basis of individual patient risk factors. The models we developed predict the probability a patient who receives radiographic imaging will have metastatic cancer. The models were developed using observational data for patients diagnosed with prostate cancer. We experimented with a variety of machine learning methods and compared their performance at predicting outcomes of imaging. The models were validated using statistical methods based on bootstrapping and subsequent evaluation on out-of-sample data. These models were used to design guidelines that seek to optimally weigh the benefits and harms of radiological imaging for detection of metastatic prostate cancer. The Michigan Urological Surgery Improvement Collaborative, a state-wide medical collaborative, implemented these guidelines, which were predicted to reduce unnecessary imaging by more than 40% and limit the percentage of patients with missed metastatic disease to be less than 1%. The effects of the guidelines were measured post-implementation to confirm their impact on reducing unnecessary imaging across the state of Michigan.

                              Co-authors:
                              Selin Merdan, Christine Barnett
                              Department of Industrial and Operations Engineering, University of Michigan, Ann Arbor, MI
                              James E. Montie, David C. Miller
                              Department of Urology, University of Michigan, Ann Arbor, MI,
                              Michigan Urological Surgery Improvement Collaborative, Ann Arbor, MI

                              Watch the Predictive Models for Optimizing Imaging Decisions for Detection of Metastatic Prostate Cancer webinar recording


                              Model Serving and MLOps

                              Speaker: Dr. Diego Klabjan, Northwestern University and Center for Deep Learning

                              Advanced analytics, in particular deep learning and real-time decision making, is ready to take the next step to transition to production and operations where MLOps and DevOps become crucial. Model serving, one of the most important part of MlOps, possesses unique challenges not present in training. We will discuss differences and challenges, and common solution principles and frameworks. You will learn the concepts of drift detection, monitoring, and feature stores. IoT use cases will be discussed

                              Watch the Model Serving and MLOps webinar recording


                              Optimization: It’s Math and Psychology

                              Speaker: Jack Levis, retired

                              Turning data into a business advantage through optimization is the goal of most organizations. The ever growing availability of data, along with expanding computing power and tools, opens the door to businesses gaining competitive advantage through analytical processes and skill. But, there is much more to making an impact than just creating advanced algorithms.  Organizations change is also essential.

                              Watch the Optimization: It’s Math and Psychology webinar recording


                              Media Manipulation in the 2020 Elections: Implications for National Security

                              Speaker: Douglas A. Samuelson, InfoLogix, Inc.

                              Elections have become increasingly driven by images, memes and social media messaging rather than policy statements.  We review some prominent examples from US Presidential campaigns and discuss how confusing and disruptive messaging was used.  We note how these methods could also be used to disrupt defense and intelligence networks and draw some lessons learned about how to increase national security.

                              Watch the Media Manipulation in the 2020 Elections: Implications for National Security webinar recording


                              Analytics for Infectious Disease Control

                              Speakers: Dr. Pinar Keskinocak, Georgia Institute of Technology and Dr. Julie Swann, NC State University

                              There has been a lot of progress in medicine and science in understanding and preventing infectious diseases over time; however, these diseases still threaten human health. A variety of prevention or intervention measures may be available, and it is important to understand which combination of these measures would be most impactful, when and where, particularly under resource limitations.  In this presentation we will provide examples on how operations research and analytics methods can  be used in supporting public health interventions for infectious diseases.

                              Watch the Analytics for Infectious Disease Control webinar recording


                              Supply Chain Resilience Through Planning and Design

                              Speakers: Zahir Balaporia, CAP, FICO Xpress Optimization and Duncan Klett, Kinaxis, Inc.

                              The COVID pandemic demonstrated, yet again, the importance of supply chains and their susceptibility to disruption.

                              This webinar will cover processes and technology for measuring and improving resiliency, as well as mechanisms  that allow organizations to respond with confidence in re-planning and execution, enabling a resilient supply chain.

                              Watch the Supply Chain Resilience Through Planning and Design webinar recording


                              Inventory, Packages, Price Points and People: An Inside Look at Leading Technical People

                              Speakers: Pallav Chhaochhria, Citi; Pooja Dewan, Otis Elevator; Mallory Freeman, UPS; Anne Robinson, Kinaxis

                              How do the best geeks in the industry motivate and manage their technical teams? Hear from:

                              • Pallav Chhaochhria, PhD
                                Director of Pricing Algorithms at Citi
                              • Pooja Dewan, PhD
                                VP, Chief Data and Analytics Officer at Otis Elevator,
                              • Mallory Freeman PhD
                                Director of Data Science and Machine Learning at UPS
                              • Anne Robinson, PhD
                                Chief Strategy Officer at Kinaxis

                              Watch the Inventory, Packages, Price Points and People: An Inside Look at Leading Technical People webinar recording 


                              Blood, Sweat, and PPEs: Rescuing Perishable Product Supply Chains and Impacting Policy through Analytics

                              Speaker: Anna Nagurney, University of Massachusetts Amherst and Supernetworks

                              In this webinar, Dr. Nagurney will talk about how this pandemic is affecting perishable product supply chains and how analytics can be used to determine the steps for recovery.

                              Watch the Blood, Sweat, and PPEs: Rescuing Perishable Product Supply Chains and Impacting Policy through Analytics webinar recording