Timeline of Developments

Below is a summary of major developments in CASTLE Labs over its history:

1983 – Developed first interactive optimization model for network design for less-than-truckload motor carriers. SuperSPIN is still in production 30 years later (marketed by Manhattan Associates). Click here for more information.

1985-1987 – Developed MicroMAP, first in-memory, real-time load matching system for truckload trucking which captured the uncertainty of the future. Still the only production load matching system for truckload 25 years later. Click here for more information.

1987 – Edelman finalist with SuperSPIN implemented at Yellow Freight System. SuperSPIN helped reconfigure the network structure for Yellow. SuperSPIN would be adopted by the entire LTL industry in the 1990’s, stabilizing this industry for the first time in the post-deregulation era.

1988 – Founded Princeton Transportation Consulting Group. Initial management team is David Cape ’87 and Ken Nickerson ’84, who jointly wrote MicroMAP.

1990 – CASTLE Laboratory founded with the hiring of Hugo Simao to develop an operational linehaul planning system for Yellow Freight System. First implemented in 1992, the system is still running 30 years later (with a major upgrade that was implemented in 2018).

1991 – Edelman finalist for application of LOADMAP (predecessory of MicroMAP) at North American Van Lines.

1992 – Developed the “LQN” methodology for fleet management, the predecessor of approximate dynamic programming for high-dimensional application.

1994 – First real-time driver scheduling system that could assign drivers over multiple legs, all using an in-memory, real-time system (developed by Derek Gittoes).

1995 – Transport Dynamics was founded by Derek Gittoes.

1996 – Belgacem Bouzaiene-Ayari joins Hugo Simao to form the core of the model development team for CASTLE Lab.

1996-2005 – Developed PLASMA for Norfolk Southern which was the first production quality optimization model for a North American freight railroad. Click here for more information.

1997 – Introduced the idea of merging historical patterns in a large-scale optimization model, blending low-dimensional rules with high-dimensional algorithms.

2002 – 10 years of research finally produced the first driver optimization model for truckload trucking which handles full set of driver and load attributes, including hours of service rules, equipment types, routing drivers to home, pickup and delivery appointment windows, home time appointments.

2002 – Greg Godfrey develops the CAVE algorithm and adapts it to resource allocation problems with multi period travel times. This logic remains one of the core tools of the lab for a wide range of applications.

2006 – Peter Frazier introduces the “knowledge gradient,” launching a new direction in optimal learning.

2007 – First edition of a book on Approximate Dynamic Programming, merging dynamic programming and math programming for the first time. Second edition (2011) was first to identify four classes of policies.

2009 – Won Daniel Wagner Prize from Informs for the first application of approximate dynamic programming for truckload fleet management for Schneider National. Now called SMART-TL, this is the first system to be able to estimate the marginal value of drivers and loads while handling all driver work rules and home constraints. Click here for more information.

2010 – Ilya Ryzhov bridges online and offline learning using the knowledge gradient.

2011 – PENSA Laboratory (Princeton laboratory for ENergy Systems Analysis) established to study stochastic optimization problems in energy, with a major grant from SAP. Click here for more information.

2011 – Second edition of Approximate Dynamic Programming appears, representing a major revision of the first edition. 300 new pages, and a complete restructuring of the book, including a first draft of identifying four major classes of policies.

2012 – Optimal Learning is published by Wiley, introducing an entirely new class of policies for information collection tuned to the needs of business, science and engineering. Click here for more information.

2012 – Developed first intercity planning system for United Parcel Service.

2013 – “CASTLE Labs” is rechristened “Laboratory for ComputAtional STochastic optimization and LEarning” to emphasize new focus on methodology with many applications.

2017 – Optimal Dynamics founded by Daniel Powell to market the “SMART” library to industry. Click here for more information.

2014-2019 – Developed new framework that unifies the “Jungle of stochastic optimization” helping to cross-fertilize tools from one community to solve problems from other communities (click here for more).

2022 – Published Reinforcement Learning and Stochastic Optimization: A unified framework for sequential decisions.  First book to span all 15 fields in the “jungle of stochastic optimization.”