ORIE’s Peter Jackson advises the winning M. Eng. project team in an annual competition.
In a competition open to the entire College of Engineering, a Systems Engineering project advised by ORIE Professor Peter Jackson has won the 2015 Silent Hoist and Crane prize.
Nearly a century ago in Brooklyn, New York, Silent Hoist & Crane manufactured a variety of heavy duty material handling equipment including forklifts, platform trucks, mobile cranes, container handlers and winches. In 1950 the company endowed an annual materials handling prize at Cornell. Although its assets were acquired in the 1990’s, the prize endures - with the scope expanded to include handling of a wide variety of materials, including information. Several of this year’s prizes were awarded at an ORIE graduation ceremony.
Systems Engineering Master of Engineering (M.Eng.) students, Anyi Xin, Lingcong Ma and Wenhan Xue, teamed up to carry out the winning project, for Canadian National Railway (CN), a long-time client for ORIE M.Eng. projects . The project dealt with a planned upgrade of CN’s pay system. Several thousand CN employees are paid based on complex formulas derived from numerous union contracts. These rules change over time, making it hard to keep payment systems up to date. So dozens of experts in these rules must double-check payroll calculations in order to catch payment errors created from the “legacy” computer system.
In anticipation of the system upgrade, CN commissioned the Cornell students to recommend a new approach for representing the rule set and to computing wages. The team designed an algorithm and built a prototype system to test it. Professor Jackson reports that CN is extremely pleased with the work, and hopes to use the prototype in a production setting until a formal replacement system can be created. To that end, CN has asked for a follow-on project this year to implement improvements in the algorithm that the winning team recommended in their project report.
Operating Room Scheduling at Cayuga Medical Center
Two of the three ORIE projects that tied for second in the prize competition dealt with healthcare operations.
ORIE students Adam Hardiman and Amelia Radtke (at right with Professor Emeritus Jack Muckstadt and Dr. Caggiano) and Michael Tillman, analyzed the policies and processes governing the scheduling of operating rooms at Cayuga Medical Center at Ithaca, a hospital that services four counties in central New York State. The project was advised by Professor Muckstadt and ORIE Professor Jim Dai.
The team spent time in the operating suite, shadowing nurses, patients and support staff and conducting interviews at each step in the scheduling process. Complementing this approach, they analyzed anonymous records of more than 9000 surgeries that were conducted over a period of more than two years. They were able to compare scheduled and actual operation durations and to look at the frequency and length of delays. They recommended ways to improve the accuracy of schedules, in particular by having surgeons make a simple classification rating of the complexity of each of their cases as they are scheduled. They also analyzed the information that flows among the staff in advance of a surgical procedure. They learned that data from surgeons’ offices can be incomplete or inconsistent, and recommended improvements in the way in which it is compiled and communicated.
David Evelyn, Medical Director of Cayuga Medical Center, said that “the students followed an iterative approach and after each meeting with me revised their data and approach, digging deeper into the data to provide new insights and ways to present the data. They also presented some usable suggestions for improvement that we are adopting.”
For team-member Amelia Radtke, “the biggest thing that I got from this experience was working on a project with a long life cycle and learning how to work with stakeholders. Another great thing about the project was taking the feedback and ideas generated in each intermediary presentation and using it to refine our final result.” She was particularly impressed with the way in which the surgeons reacted to the team’s idea of a complexity rating, “feeling ownership of the idea before buying into it.” Project advisor Muckstadt noted that “I often tell students that obtaining buy-in requires the client to personally own the idea. That is what ultimately happened. Hence the students had an experience that reflected one of my key principles of effective consulting.”
Radtke, who is from Milwaukee, WI and a graduate of ORIE’s undergraduate program, now works at Froedtert & the Medical College of Wisconsin, a system of three hospitals serving southeastern Wisconsin, in a job she says would have been unavailable to her without her project work. She notes that “While most people in health care recognize the importance of my skills and background in moving the industry forward, people are still wary of my atypical background. However I was able to cite my project work to demonstrate how my degree is applicable, and that I am able to learn and adapt to working in health care.’ She has also found her experience in ORIE’s courses in simulation, systems engineering, spreadsheet-based modeling and data analysis, and application of OR to health care to be strong contributors to success so far in her chosen profession.
The Hospital for Special Surgery: Analysis of Block Schedule and Central Sterile Processing Operations
Professor Muckstadt also advised another of the 2nd place winning projects, the second of a series to improve the processes for cleaning, sterilizing and preparing surgical instruments for the operation rooms at a top orthopedic and rheumatology hospital in New York City , the Hospital for Special Surgery.
Team members Daniel Dworakowski, Rajeev Alluri, and Scott Wiebel (at left) worked with the Central Sterile Processing unit that employs 85-100 fulltime employees at the hospital. These employees assemble carts carrying metal trays of instruments to be used by surgeons in the operating rooms. The trays hold from one to fifty instruments, many specified by the surgeon who will perform the operation for which they are destined. Following surgery the instruments undergo a sterilization process that takes at least five hours. An average of 1200 trays is used in the average of 120 surgical procedures done per weekday at the hospital.
The team worked to fully understand the processes used in the Central Sterile Processing Unit, the usage by day and time of individual tray types, and the occurrences of excess and understocked trays, with objective of reducing effort and saving space in the facility. They built a mathematical optimization model to determine how modifications in the scheduling of blocks of operations by individual surgeons could smooth out demand for instrument trays. This led to a heuristic ‘rules of thumb’ approach to making small changes in schedule that would significantly reduce the variation in daily requirements for trays. They also recommended changes in the management of the onsite tray inventory that could reduce the investment in the inventory by millions of dollars.
Scott DeNegre, the hospital’s Assistant Vice President for Operational Excellence, said that “team demonstrated significant initiative and excitement about the project. We were so happy with the outcome of the first project [the previous year], we decided to design this year's project as a follow on to the prior work. This year's project was just as successful and I'm looking forward to sponsoring again next year!”
Team member Rajeev Alluri, whose undergraduate work at Cornell was in Biological Engineering, said that the project “was an amazing experience because we got to interact with a lot of high level individuals in the operations team of the hospital.” Alluri now works for Axtria, a data analytical consulting company in New Jersey that serves clients in the life sciences, healthcare and several other industries.
Team member Scott Wiebel, a University of Miami graduate in Biomedical Engineering, “liked working with the Operational Excellence team to see how operations research skills can be used to improve a top organization.” He is now a member of the Operational Excellence team as an intern.
Pitney Bowes Presort Services: Forecasting System
Pitney Bowes, known to many as the provider of postage meters that weigh and add postage to envelopes and packages at the point of origin, now has a much broader set of offerings to “power the transactions that drive commerce,” which products and services generate nearly $4 billion in revenue from 90% of Fortune 500 companies. One major Pitney Bowes business line, Presort Services, partners with the US Postal Service to presort mail by zip code, thereby qualifying delivery for reduced postal rates. Variation in the volume of mail picked up and delivered for processing in the company’s 35 facilities where mail is presorted is a challenge in workforce scheduling.
A team comprised of Shiying Wang, Elisa Gunawan, and Ellen Patridge (at right) under the guidance of Professor Mark Lewis and ORIE Master of Engineering Director Kathryn Caggiano - worked with the Pitney Bowes facility in Reading, PA, one of the closest presort locations to Ithaca. The main objective of the project was to create a software tool that forecasts incoming mail volume as a function of factors such as shift, day and month, weather conditions, and the pattern of customer orders.
Rather than focus on records generated by the presort machines, which do not necessary reflect the rate at which mail has arrived for processing, the team obtained daily logs of the arrival of mail at the dock. They used a technique known as linear least squares regression modeling to explain the contribution of each factor to the variation in workload. They embedded this technique in easy-to-use software built on Microsoft Excel and an add-in product called RealStats so it can be used by Pitney Bowes planners, for whom they provided a user guide.
The team was able to use their software to show the importance of enhancing communication with a small number of high impact customers. Knowing more about the load created by just a few customers increased the ability of the model to forecast load and account for as much as 80% of the variability. They thereby offered the possibility that using demand forecast information could give employees more certainty about scheduling while reducing the costs associated with paying more expensive temporary staff.
Bruce Gresham, Vice President of Business Planning and Strategy for Pitney Bowes Presort Services, said “the project has the additional support of our business unit President as well as the Vice President of Operations. I expect it to be implemented.” He was particularly complimentary towards the students, who “acted as a cohesive, thoughtful team” and showed “a depth of understanding of our business that is impressive given their short time working with us. They are ready to be junior leaders within a company.”
For team member Ellen Patridge, “the most exciting part of the project was making the final presentation to the company.” The presentation “created discussion among the leadership present, and gave us hope for significant changes for how they view mail arrivals in the future.” Patridge, from Macedon near Rochester NY, started her Master of Engineering program while still an ORIE undergraduate, and will receive her M.Eng. degree at the end of 2015.
Elisa Gunawan, who is from Jakarta, Indonesia, earned a Bachelor’s degree in Engineering and a Master of Entrepreneurship at the University of Michigan before coming to Cornell. For her “the highlight of the project was working with my teammates, because they truly are amazing people both to be workmates and friends with,” she said. She also noted that “it is great to be able to gain insights into the business model as well as the operational aspects of a massive company like Pitney Bowes.”