DESCRIPTION

Amazon's Middle Mile Planning Research and Optimization Science group (mmPROS) is looking for senior research scientists specializing in the development of forecasting, data science and stochastic optimization algorithms applied to transportation planning and yield management. This includes the development, enhancements and implementation of forecasting models and systems, and creating analytical tools to improve transportation planning solutions.

Middle Mile Air and Ground transportation represents one of the fastest growing logistics areas within Amazon. Amazon Fulfillment Services transports millions of packages via air and ground and continues to grow year over year. The scale of this operation challenges Amazon to design, build and operate robust transportation networks that minimize the overall operational cost while meeting all customer deadlines. The Middle Mile Planning Research and Optimization Science group is charged with developing an evolving suite of decision support and optimization tools to facilitate the design of efficient air and ground transport networks, optimize the flow of packages within the network to efficiently align network capacity and shipment demand, set prices, and effectively utilize scarce resources, such as aircraft and trucks. Time horizons for these tools vary from years and months for long-term planning to hours and minutes for near-term operational decision making and disruption recovery. These tools rely heavily on forecasting algorithms, mathematical optimization, stochastic simulation, meta-heuristic and machine learning techniques. In addition, Amazon often finds existing techniques do not effectively match our unique business needs which necessitates the innovation and development of new approaches and algorithms to find an adequate solution.

The Sr. Principal Scientist will have a key role in Amazon's transportation system's strategic planning and operation by successfully partnering with various science, engineering, operations, and analytics teams to develop forecasting models as a crucial component of Amazon's middle mile optimization and planning tools. The candidate will impact Amazon's business by driving continuous improvements in forecast accuracy, and better understanding and managing transportation demand volume variance drivers.

The candidate will work closely with Amazon leadership and the rest of the operations research and data science teams to leverage the expertise of each individual to construct models, perform analyses, and derive relevant metrics. The candidate must have relevant domain knowledge to teach and mentor group members and to critique models and approaches taken by the group in terms of business relevance, technical validity, software architecture, and computational performance. The candidate must have the skills to write documents that influence important investment and resource allocation decisions by clearly articulating the strategy, business impact, and technical challenges.

BASIC QUALIFICATIONS

• Ph.D. in Operations Research, Statistics, Applied Mathematics, Computer Science or a related field with publications in refereed academic journals.
• At least 15 years of experience in solving complicated optimization and machine learning problems for transportation networks or analogous disciplines developing a strategy for large-scale networks.
• Experience designing and implementing transportation optimization models with focus on volume and route planning.
• Excellent communication skills, both written and oral with both technical and business people. Ability to speak at a level appropriate for the audience. Experience applying these skills in both academic teaching environment and a business setting is a plus.
• Excellent writing skills for presenting business cases and to document the models and analysis and present the results/conclusions in order to influence important decisions.
• A working knowledge of smooth and non-smooth optimization methods accompanied by associated expertise in the use of tools and the latest technology (e.g. CPLEX, Gurobi, XPRESS).
• A working knowledge of exact, approximation algorithms, and heuristic methods for solving difficult optimization problems like vehicle routing and network design problems.
• The ability to implement models and tools through the use of high-level modeling languages (e.g. AMPL, Mosel, R, Matlab).
• Experience prototyping and developing software in traditional programming languages (C++, Java, Clojure, Python).
• Familiarity with SQL and experience with very large-scale data. The ability to manipulate data by writing scripts (Perl, Ruby, Groovy) a plus.
• Statistical analysis, machine learning and data-modeling in a database environment is a plus.

PREFERRED QUALIFICATIONS

Professional traits that are not unique to this position, but necessary for Amazon leaders:
• Exhibits excellent judgment and relentlessly high standards
• Thinks strategically, but stays on top of tactical execution
• Expects and requires innovation of her/his team
• Thinks big and has convictions
• Results oriented
• Has the innate ability to inspire passion in others