Welcome

Welcome to my blog that is intended to keep interested parties up to date on my latest research and teaching endeavors.  Specifically, I view this blog as a way to engage the online community by:

  1. Sharing — As we live in a society where the majority of us spend a great proportion of our days “staring at glowing rectangles” – I thought a digital presence to share my current research findings, projects, and insights would be valuable to the academic and industry community.
  2. Engaging — If anything you see on my blog is of interest to you, please contact me through email.  I am constantly looking for interesting research projects that are motivated by industry problems, as well as research collaborators throughout the world.
  3. Promoting —  Industrial engineering is a profession I am very lucky to have discovered and is a profession I find extremely valuable, practical, and rewarding.  Unfortunately, it is also a profession that tends to have a marketing problem.  When I give presentations to prospective students, I constantly get comments such as “I had no idea an industrial engineer did that.”  If this blog, in some small way, can get the word out about industrial engineering, there is an opportunity to match people with a career they find equally rewarding.
Advertisements

Post-Doctoral Research Position Available in Our Research Lab

A post-doctoral position is available in Rensselaer Polytechnic Institute’s Industrial and Systems Engineering Department.  In this appointment, you will work under the supervision of Dr. Jen Pazour to research new ways to meet the demands of modern distribution.  The ideal candidate will have a strong background in optimization methodologies, ideally (but not required) with some exposure to bi-level optimization and Stackelberg games.  The initial appointment is one year, with potential for renewable based on satisfactory performance and funding availability.  The position is available immediately.    The ideal candidate would start by September 1st, 2018; while this starting date is flexible, priority will be given to candidates with earlier start dates.

If interested, please apply to the open post-doc position at this link: https://rpijobs.rpi.edu/postings/6639 and send your CV to Dr. Jen Pazour at pazouj@rpi.edu.

More information about our group’s research can be found here: https://jenpazour.wordpress.com/.  This specific post provides details about the types of skills ideal for this open position: https://jenpazour.wordpress.com/2017/12/11/looking-for-new-ph-d-students-to-join-my-research-lab/   Feel free to contact Jen (pazouj@rpi.edu) if you have questions or to request a SKYPE appointment to learn more.

Efficiency on Demand

Article written by John Backman for Rensselaer’s ISE Spring 2018 newsletter

Jennifer Pazour is finding ways to streamline supply chains in the new collaborative economy

Jennifer Pazour started early on her quest for optimal efficiency.

“I really enjoyed organizing my room as a kid,” said the ISE assistant professor. “I have just always liked efficiency.”

The lifelong fascination recently led to two of engineering’s most prestigious awards—and perhaps a better life for Uber drivers, Meals on Wheels volunteers, and others who inhabit on-demand supply chains.

Early this year, the National Science Foundation awarded Pazour a Faculty Early Career Development (CAREER) grant. More recently, Johnson & Johnson honored her with its Manufacturing Scholar Award, given as part of the WiSTEM2D Scholar Program (Women in Science, Technology, Engineering, Mathematics, Manufacturing, and Design).

The research behind the accolades has to do with Pazour’s signature focus: the seismic shift from centralized to on-demand and collaborative distribution.

“Modern distribution systems need to fulfill a wide variety of requests quickly with little warning in small units to many dispersed locations at low costs,” she wrote in her research summaries.  “This is fundamentally different than yesterday’s demand, which aggregated at fixed (store) locations.  Thus, today’s supply chains are optimized for yesterday’s customers.”

In the face of this disconnect, Pazour’s team is rethinking supply chain design to meet the demands of modern distribution.  That involves researching new ways to move supply chain networks from fixed and static to collaborative, dynamic and agile.

One aspect of this research involves underutilized resources and how organizations might obtain them.  To understand it, consider cars and what they do – or don’t do – all day.

“The tings we own have extremely low utilization rates, spending the majority of their useful lives idle,” Pazour explained.  “When you’re at a stoplight, count the empty seats in the vehicles around you.  When you’re in a parking lot, think about the fact that most of the surrounding cars get an hour of use a day.  Or consider the duplication of effort when both you and your neightbor make invividual grocery trips.  These example all represent underutilized capacity, and with the right algorithms and models, companies could start accessing these underutilized resources.”

But there’s a challenge in the way.  “We need to entice the owners of these resources to provide access,” she continued.  “On the one hand, the centralized model doesn’t allow for decision making by the owners, so it dampens their participation.  On the other, a fully decentralized system leads to myopic decision making: no one supplier has the whole picture of the marketplace, which results in reduced system performance because some requests receive multiple selections and others are left unfulfilled.”

The CAREER award-winning research aims to combine the two in what Pazour calls a hierarchical approach, which gives suppliers “recommendations” to help them make efficient decisions. She used Meals on Wheels volunteers as an example.

“Millennial Millie gets a notification on her phone asking if she’d like to deliver groceries to shut-in residents,” she recounted. “Millie clicks yes. Two requests appear. She chooses the one that fits with her plans that day. What the platform did—without control or knowledge of Millie’s plans—was to provide choices estimated from her past behavior.  Before this example can become a reality, research is needed to discover new ways to provide choices and quantify the impact of those choices on suppliers and demand requests.”

Another part of the Johnson & Johnson and CAREER research dovetails with Pazour’s dedication to her profession. Working with Rensselaer’s Engineering Ambassadors program, she will mentor undergraduates to create active learning modules, inspired by her research, and use them with K-12 students.

“I want to gain more exposure for the field, and inspire more young people to get involved,” she said. “The world need more industrial and systems engineers. We are wired to think systematically about complex problems, which exist in all sectors of society. Yet too few incoming Rensselaer students even know what we do. That needs to change.”

 

Aside
IMG_2806

Zach Shearin, Presenting his Research at the 2017 IISE Regional Conference

Zach Shearin, an undergraduate Industrial and Management Engineering student at Rensselaer Polytechnic Institute and a die-hard Carolina Hurricanes fan, used analytics to analyze the National Hockey League’s point system through an operations research and statistical analysis lens.  This work started as Zach’s project for my Operations Research Methods (ISYE 4600) course.  He then continued research and analysis as an undergraduate researcher.  The NHL’s current point system awards more total points to games ending in overtime than games ending in regular time.  In this work, we evaluate if this a fair system, and if it influences style of game.  We find that the current point system results in statistically more passive play in the last five minutes of regulation for an even-score game than the first fifty-five because both teams want to ensure at least one point. The “3-2-1-0” point system minimizes discrepancies with win-loss record, and does not compromise the competitiveness and entertainment of the game and is recommended.  Zach presented this research at the 2017 Regional Institute of Industrial and Systems Engineering Conference, which earned him a second place finish.  We have submitted this work to the 2018 IISE Conference; our submitted conference paper can be downloaded here: [Shearin and Pazour, NHL Point System Fairness Optimization and Game Play Stat Analysis].

NHL

Click here for the full analysis and post

 

 

2017 Recap

Welcome to our 2017 Recap!  These Year-in-Review posts, an annual tradition, catalog our team’s progress while encouraging reflection and preparation for the voyage into the new year.

Quote: Rainer Maria Rilke #riflepaperco

Quote by Rainer Maria Rilke, Artwork by Rifle Paper Co. 

2017 culminated in two success stories. Dr. Uzma Mushtaque defended her PhD Dissertation entitled Context Dependent Discrete Choice Models and Assortment Optimization for Online Retail. Joan Climes defended her MS Thesis on Analytical Models for Retrieving Items in Dense Storage Systems and Optimizing the Location of an Open Square.

Uzma developed descriptive mathematical models to capture context-effects associated with individual user selection behavior.  Her models are novel as they capture the influence of assortment properties (specifically the assortment size), in addition to user and item attributes (as are commonly captured in existing research).   Using her new class of random utility models as inputs to optimization problems, she proves insights and creates new algorithms to determine “what to recommend” and “how many to recommend” in online settings.  She validated her approaches using data from online movie recommender systems and online retail. More details are available in this paper under review.  This research, partially supported by the National Science Foundation, earned Uzma an honorable mention at the 2017 IISE Doctoral Colloquium and a trip to Amazon’s Graduate Research Symposium. Uzma is currently serving as a Post-Doctoral Researcher in my lab and an instructor for the Core-Engineering course Modeling and Analysis of Uncertainty.

Joan’s research focused on dense storage systems, which allow for highly effective use of space, at the expense of requiring the repositioning of stored items to retrieve other more densely desired items. These dense storage systems are found in warehouses and distribution centers, and aboard US Navy ships used for sea-based logistics (this work was partially supported by the Office of Naval Research).  Her research creates mathematical models to determine the value of an empty space in a specific dense storage environment, the double-inverted T configuration (discovered by the always innovative Kevin Gue). Retrieval and repositioning distance equations are derived for each item in a layout.  An optimization problem is presented to select which location should be left open. Best locations for an open square are along the aisle and close to the vertical walls if h > k, or close to the horizontal wall if h = k. Due to the symmetry in repositioning distances, multiple optimal solutions exist.  Joan has accepted a position with Deloitte starting in February. In the meantime, we hope to work together to submit this work for peer-review.  Joan conducted this independent research as an undergraduate student, leading the modeling and coding development, all while taking challenging PhD-level optimization and statistics classes. She happens to be a runner on the RPI track team too. She has a standing offer to rejoin our team and pursue a Ph.D., as do a number of talented undergraduate researchers I’ve been honored to work with in the past.

Due to Uzma and Joan’s graduations, my research group has openings. I am looking for curious, talented people to join my research team.  If this year doesn’t work, keep us in mind in the future.

Kaan Unnu made great research progress in 2017 for his dissertation “Optimization Models for On-Demand Supply Chain Collaboration.”  On-demand systems provide resource elasticity: enabling finer granularity capacity and commitment decisions, and access to scale.  Kaan has chosen on-demand warehousing as a focus. Novel mixed integer linear programming models and efficient solution algorithms decide location-allocation in a dynamic network, capturing build, lease, and on-demand distribution simultaneously.  Computational experiments, utilizing the mathematical models, identify significant factors impacting performance and codify policy recommendations.  We’ve also partnered with IBM Research to start exploring the potential for blockchain technology to improve trust and facilitate physical movement/storage of goods’ data into a distributed ledger system.

Shahab Mofidi defended his PhD candidacy this summer, which focuses on “agile resource allocation decisions in modern supply chains with on-demand suppliers”.  Most recently, he has been developing new models and algorithms for online platforms.  To understand the trade-offs of providing choices to drivers through simultaneous personalized recommendations, we propose a hierarchical decision-making framework where the platform decides a recommendation set for each driver. Drivers then have discretion to choose the riders that best fit with their preferences or planned travel from this set. We model this framework as a bilevel optimization problem with a profit maximizing objective for the platform in the upper level (leader) problem and a utilitarian social welfare objective for the lower level (follower) problem. This results in a computationally expensive mixed integer linear bilevel problem. Since the platform needs to make instant recommendation for a relatively large problem size, we transform the formulation into a single level problem through proposing logical expressions. This research provided preliminary results for a NSF research proposal I submitted this summer. Shahab is a crucial resource.  He’s an excellent collaborator, and he continues to win national awards and scholarships.

IISEAward.png

I was humbled to be awarded the 2017 IISE Dr. Hamed K. Eldin Outstanding Early Career IE In Academia Award.  This award is especially exciting to me because so many of the past recipients are human beings whose careers I admire and hope to emulate.

Pazour_SDSMTGraduation

I was awarded the 2017 SDSM&T Outstanding Recent Graduate.  As a blast from the past, I dug out my undergraduate graduation speech, which included the lines below…with a few inside jokes.

You might be a Tech grad if…

You can name the flavor of the day at Armadillos for the whole next week.

When asked to take a picture, you count off 1, 3, 5

You’ve ever applied probability and statistics in Deadwood or fluid dynamics to a night on the town.

Your student ID is only important one day of the week and that’s Wednesday.

You actually know your professors and they know you.

In 2017, I enjoyed giving back to my graduate school Alma mater – the University of Arkansas – by serving on their IE Liaison Board.  The IE department is conducting innovative research, while keeping students the focus.

We wrapped up our work on Sea-Based Delivery systems, funded through the Young Investigator program by the Office of Naval Research.  2017 saw two papers accepted for publication out of this research, with a couple more in the pipeline.  I joined the editorial board of IISE Transactions.  Having handled my first paper as an associate editor of the Emerging Applications and Analytics Department, I was reminded of the flip-side of the peer-review process.

2017 provided plenty of evidence the future is bright.  This included undergraduate ISE majors from across the Northeast giving up their weekend to attend the 2017 IISE Regional Conference, hosted at RPI.  Zach Shearin did a great job presenting his research on Analytics for the NHL Point System, earning him 2nd place in the undergraduate technical paper competition.  The winning Rutgers team did a fantastic job, and earned 2nd place at the National Competition. I was inspired by research conducted by undergraduates across the globe as I served on INFORMS Undergraduate Operations Research Prize committee.  I am the chair for the prize committee in 2018, and look forward to receiving inspiring applications.   Serving as RPI’s IME Class of 2020 adviser, our students’ focus, self-awareness, and vision, which are much beyond what I was thinking about as a freshman/sophomore in college, are encouraging.

Teaching brings me great joy.  I enjoyed polishing my course materials for two courses in 2017: Design and Analysis of Supply Chains and Operations Research Methods.  I am a big fan of clicker questions to keep students engaged and participating in the materials.  I’ve found these can work for quantitative materials too.  Please click in:

D3 Inventory Management Uncertain Demand.jpg

I spent a lot of time in the Fall preparing to give a RED (Research, Education, and Discovery) Talks – A Transformative Rensselaer Confronts the Global Challenges.  My presentation with Professor Jim Hendler was about “The Data Challenge.”  I presented my vision of the future of supply chains and how researchers across RPI are addressing the need for tomorrow’s supply chains to be resilient and agile.

Our research was featured in a 2017 Supply Chain Dive article “How retail supply chains are adapting to the Amazon effect.”  In particular, we were quoted, “A wide variety of our [order] requests are made with very little warning and are expected to be fulfilled quickly, in small units, to a number of different locations,” Jennifer Pazour, Assistant Professor of Industrial and Systems Engineering for Rensselaer Polytechnic Institute said at a WERC conference panel. This is the Amazon effect. This is the idea that I want my stuff now and fast, and oh by the way I don’t want to pay much for it.”  This is fundamentally different than yesterday’s demand, which aggregated at fixed (store) locations.  Thus, today’s supply chains are optimized for yesterday’s customers. To close the gap between current supply chain operations and customer expectations, our team rethinks supply chain design.   Our research was well-received by industry groups, having presented in 2017 at three separate APICS events, at WERC’s annual meeting and through WERC’s webinar series.  Such interactions have led to connections and research projects with supply chain and logistics companies and start-ups.  I’ve also continued involvement with a number of programs to encourage youth to pursue a career in engineering and logistics.  A highlight was the presentation given by undergraduates Brook Rulewich and Fiona Flynn, created as part of RPI’s Engineering Ambassador’s program.  Their presentation, geared toward innovative ways to deal with traffic, motivates middle and high school students to think about careers in Industrial and Systems Engineering, and Civil Engineering.

Collaborations created through the Gulf Research Program Fellowship program have led to new friends, new problems, and new data.  I am specifically excited about my collaboration with Diego Figueroa, from the School of Earth Environmental and Marine Sciences, University of Texas Rio Grande Valley, to explore Optimal Marine Protected Area Design for Mesophotic Reef Conservation in the Gulf of Mexico.

To avoid the trap of social media only showing a curated life of success, I should discuss some disappointments in 2017 as well. I received the gut-wrenching email that my proposal has been rejected 4 times, along with several similar emails for paper submissions.  As I mentioned in a presentation to the IISE Doctoral Colloquium, a research career is not a monotonically increasing function: feedback and criticism lead to better end products and growth.

Pazour Doctoral Colloquium

But 2017 wasn’t all research and papers! In 2017, I marched for science, equality, and facts.  We become first-time homeowners, saw 90% of the solar eclipse, visited the Math Museum, celebrated the successful return of Crash Bandicoot (and his sister Coco), enjoyed the consistency of meeting up with friends at Troy’s Farmer Market and Wine Wednesdays, enjoyed get-aways to MASS MoCA, hosted family in Troy, read a few great books…and a few ok books.

 

Cheers to 2018!

Jen

Looking for new Ph.D. Students to Join My Research Lab

RED Talk Hendler and Pazour FINAL

Modern distribution systems need to fulfill a wide variety of requests quickly with little warning in small units to many dispersed locations at low costs.  This is fundamentally different than yesterday’s demand, which aggregated at fixed (store) locations.  Thus, today’s supply chains are optimized for yesterday’s customers.

To close the gap between current supply chain operations and customer expectations, our research team rethinks supply chain design.  We are exploring creative solutions around on-demand warehousing and crowdsourced deliveries, in which marketplaces provide access to resources (when and where they are needed), rather than owning them.  This creates a dynamic supply network, able to respond to changing demand requirements.  But, such systems are inherently more complex than traditional systems. To address these challenges, we are researching new network design models to capture on-demand business models (for a quick overview,  check out Flexe’s videos ) and will use these models to quantity the benefit in terms of access to scale, reduced commitment granularity, and reduced capacity granularity.  We are also conducting basic research on how best to provide a set of decentralized suppliers choice to entice them to provide access to their resources on-demand. By tapping into underutilized supply capacity, supplier choice can increase participation – and thus capacity –  and provides agility through more flexible use of suppliers.  This can improve e-commerce profitability and enable a new on-demand volunteer base.  Our research team has partnered with community nonprofits to test how on-demand grocery delivery systems for mobility-restricted clients can help address the needs of residents living in food deserts.

Sound interesting?  If so, I encourage you to apply to join my research lab.  I’m looking for talented, curious new Ph.D. students to join in Fall 2018.   Check out my website (https://jenpazour.wordpress.com/) to learn more about our research and team.  Feel free to contact me (pazouj@rpi.edu) if you have questions or to request a SKYPE appointment to learn more.

Earning a Ph.D. fundamentally changed — for the better — both my professional and personal paths in life.  I encourage you to think about a Ph.D. I also entered the Ph.D. program pretty naive.  So below I provide what I (now) believe is needed to be successful in a Ph.D. program, and a research career beyond, as well as reasons why I think RPI’s Industrial and Systems department is a good place to be a Ph.D. student.

First, a Ph.D. is a research degree.  This is in contrast to bachelor degrees and (today) most masters degrees, which are coursework degrees.  Getting good grades in school is not sufficient to succeeding in graduate school.

So what do I believe it takes to succeed in graduate school?  Well, first, you have to be ready to fail.  Research is about discovering something new or doing something that has never been done before.  There are no answers at the back of the book.  The discovery process is exciting, but also non-linear.  Many of the things we try, do not work.  You have to be OK with this.

But, you also need to succeed enough to outweigh all the failing.  Bob Dylan, the great Nobel prize-winning poet, summed up life in academia well, “She knows there’s no success like failure.  And that failure’s no success at all.”  You need to be excited about what you are doing and willing to put in the time and follow through the failures to get to success.  Because ultimately to graduate, you need to succeed.  Your research needs to make contributions.  Failure is not enough.  Follow through is critical.  You need to be able to make yourself do the mundane (whether that’s writing up results, responding to reviewers comments, writing up research funding reports, or responding to emails).  In fact, to be successful as a faculty member, I believe you need to be efficient at the mundane.  You need to be able to efficiently juggle many different projects, requests, and emails.

To succeed at a Ph.D. it is (in my opinion) necessary to be able to do both: to (1) excel at new idea generation and to bounce back after failure, and to (2) follow through (and even be efficient) with mundane tasks.  While you do not need to an expert at either of these as an incoming Ph.D. student,  you need to work at and keep improving on both types of tasks.  Therefore, skills I am particularly looking for in Ph.D. Students:

  1. Helpful Human Being
  2. Grit
  3. Kindness
  4. Curiosity
  5. Writing and Logic Skills
  6. Communication Skills
  7. Ability to think about and improve our understanding of complex problems
  8. Mathematical ability
  9. Coding background
  10. Ability to deal with uncertainty.
  11. Critical Thinking.

Reasons why I would like to be a graduate student at Rensselaer’s Industrial and Systems Engineering department:

  1. It’s a small co-hort. We have a small, but mighty faculty, and our Ph.D. admission process is deliberately geared towards recruiting a small, but mighty group of Ph.D. students. We limit the number of Ph.D. students admitted to ensure each Ph.D. student fits with the research interests of our department and is provided mentoring and funding.
  2. We are in this together. Research in my lab is a collaborative process.  This means my graduate and undergraduate students and I are putting our heads collectively together to generate new knowledge and create new models and methods.  We work on the research together.  I meet with Ph.D. students at least once a week and spend many additional hours reading, editing, and writing journal papers with my students.
  3. It’s a disruptive time to be in supply chain design. As the introduction to this post illustrates, its a disruptive time to be in supply chain design.  My hypothesis is that supply chains of today are optimized for yesterday’s customers.  This means the world needs more creative ideas and needs to utilize the massive amounts of data being generated today to drive decision making.  Thus, the research we are doing is important and has the potential to improve efficiency of commercial and nonprofit organizations.  To learn more, check out my RED (Research, Education and Discovery) Talkthe Data Promise” (where I describe research on data in supply chains starting at 7:50), a webinar where I describe on-demand warehousing and logistics, and this interview about on-demand logistics .
  4. Low boundaries to collaboration across campus. My students and my research have benefited greatly from the insights of RPI’s helpful faculty members, both in the ISE department and beyond. Ph.D. students are encouraged to take courses across disciplines (e.g., Ph.D. students take data analytics from IDEA, optimization theory from mathematics, machine learning, data mining, and algorithms classes from computer science, transportation and econometric modeling from Civil Engineering, queuing from Electrical Engineering, data analytics from ITWS, sourcing from Lally business school, and decision making from cognitive science departments).

 

Applicants are encouraged to apply to Rensselaer’s Decision Sciences and Engineering Systems Ph.D. program, housed in the Industrial and Systems Engineering Department.  Please indicate in your statement of purpose, your interest to work with me.  If you currently reside in the US, we sponsor trips to have top accepted PhD students visit the campus, meet with faculty, and see the area.  Students in my research lab are funded, either via research assistantships, teaching assistantships, or fellowships.  Additional funding and scholarships are available for talented domestic applications.

If interested, please apply by January 2nd, 2018.   Note, applications are reviewed beyond the deadline, so apply even if you miss the January 2nd deadline.  Admission decisions are made by a committee and the final decision resides with the graduate school (outside of our department).

Please reach out to me via email (pazouj@rpi.edu) if you have any questions.  Tell your friends, co-workers, former students, current students, etc.  Thanks! Jen

On-Demand Warehouses

Check out the latest WERC Sheet for a nice write-up about on-demand warehouses.

WERC_Sept-Oct2017_web_Page_04

WERC_Sept-Oct2017_web_Page_05.jpg

My research team and I are rethinking supply chain design.  We are exploring creative solutions around on-demand warehousing and crowdsourced deliveries, in which marketplaces provide access to resources (when and where they are needed), rather than owning them.  This creates a dynamic supply network, able to respond to changing demand requirements.  But, such systems are inherently more complex than traditional systems.  To address these challenges, we are working on new network design models to capture on-demand business models (for a quick overview,  check out Flexe’s videos ) and will use these models to quantity the benefit in terms of access to scale, reduced commitment granularity, and reduced capacity granularity.  We are also conducting basic research on how best to provide a set of decentralized users choice to entice them to provide access to their resources.

Blockchain Technology – Guest Blog Post

blockchain

Happy (almost) End of the Semester.  I am happy to share a guest blog post by two undergraduate researchers, Mara and Jake, about their adventures into blockchain technology.

I’m hoping to get more undergraduate students blogging about their research interests, and so added them under the people heading.  Click here for Mara and Jake’s blog post about blockchain technology for supply chain applications.

I’m super excited about this technology and think it has great potential in supply chains.  Specifically, I believe supply chains and logistics are becoming more democratized, which results in a need for decentralized and distributed decision making.  Another area of my research has been in improving supply chain visibility.  In theory, visibility should be solved by now, but I have a number of data points that show in practice it isn’t.  My hypothesis for this discrepancy is that a top-down approach to visibility is hard to achieve in practice.  This is because to have visibility into your supply chain requires decentralized entities to agree to connect data bases and systems, share data, etc., and this is a challenging management and technology problem.  One promising technology to improve visibility is blockchain technology, which uses decentralized protocols capturing and validating information transactions between multiple users via a distributed ledger system.  My hope is to report more on this research in future posts.