Optimal Portfolio Strategies for New Product Development

We study the portfolio selection problem in a new product development setting with many projects in parallel, each lasting several stages, in the face of uncertainty. Each stage of the process performs an experiment on a selected number of projects in the stage, depending on the amount of (scarce) budget allocated to the stage. Projects become differentiated through their experimental results, and all available results for a project determine its category.We model the problem as a Markov decision process. We seek an optimal policy that specifies, for every configuration of projects in categories, which projects to test and/or terminate. For two special cases we characterize the optimal project promotion policy as following a new type of strategy, state-dependent non-congestive promotion (SDNCP). SDNCP implies that a project with the highest expected reward in any stage is advanced to the next stage if and only if the number of projects in each successor category is below a congestion-dependent threshold. For the general problem, numerical experiments reveal the outstanding performance of SDNCP (optimal in 72 of 77 instances with maximum deviation from optimal of 0.67%), highlighting when and how a fixed non-congestive promotion policy, which is easier to implement, may fall short.

[1]  Albert Y. Ha.  Optimal Dynamic Scheduling Policy for a Make-To-Stock Production System , 1997, Oper. Res..

[2]  Stefan H. Thomke,et al.  Six Myths of Product Development , 2012 .

[3]  Ming Ding,et al.  Structuring the New Product Development Pipeline , 2002, Manag. Sci..

[4]  P. Whittle.  Restless Bandits: Activity Allocation in a Changing World , 1988 .

[5]  Lawrence M. Wein,et al.  Dynamic Scheduling of a Multiclass Make-to-Stock Queue , 2015, Oper. Res..

[6]  Karl T. Ulrich,et al.  Product Design and Development , 1995 .

[7]  Robert G. Cooper,et al.  Perspective: The Stage‐Gate® Idea‐to‐Launch Process—Update, What's New, and NexGen Systems* , 2008 .

[8]  Christoph H. Loch,et al.  Accelerating the process of engineering change orders: capacity and congestion effects , 1999 .

[9]  Albert Y. Ha.  Inventory rationing in a make-to-stock production system with several demand classes and lost sales , 1997 .

[10]  Martin L. Puterman,et al.  Markov Decision Processes: Discrete Stochastic Dynamic Programming , 1994 .

[11]  J. Gittins.  Bandit processes and dynamic allocation indices , 1979 .

[12]  Alan Scheller-Wolf,et al.  Experimental Results Indicating Lattice‐Dependent Policies May Be Optimal for General Assemble‐To‐Order Systems , 2016 .

[13]  Steven A. Lippman,et al.  Applying a New Device in the Optimization of Exponential Queuing Systems , 1975, Oper. Res..

[14]  Gary Blau,et al.  Risk management in the development of new products in highly regulated industries , 2000 .

[15]  P. Whittle.  Arm-Acquiring Bandits , 1981 .

[16]  L. J. Savage,et al.  Three Problems in Rationing Capital , 1955 .

[17]  M. Weitzman,et al.  FUNDING CRITERIA FOR RESEARCH, DEVELOPMENT, AND EXPLORATION PROJECTS , 1981 .

[18]  Stylianos Kavadias,et al.  Resource allocation and new product development portfolio management , 2007 .

[19]  Sheldon M. Ross,et al.  The Streetwalker’s Dilemma: A Job Shop Model , 1971 .

[20]  J. Bather,et al.  Multi‐Armed Bandit Allocation Indices , 1990 .

[21]  Atul Bhandari,et al.  An Exact and Efficient Algorithm for the Constrained Dynamic Operator Staffing Problem for Call Centers , 2008, Manag. Sci..

[22]  M. Puterman,et al.  Bias optimality in controlled queueing systems , 1998 .

[23]  R. W. Hansen,et al.  The price of innovation: new estimates of drug development costs. , 2003, Journal of health economics.

[24]  Karl T. Ulrich,et al.  Valuing R&D Projects in a Portfolio: Evidence from the Pharmaceutical Industry , 2007, Manag. Sci..

[25]  Christoph H. Loch,et al.  Project Selection Under Uncertainty: Dynamically Allocating Resources to Maximize Value , 2004 .

[26]  Dimitris Bertsimas,et al.  Conservation laws, extended polymatroids and multi-armed bandit problems: a unified approach to ind exable systems , 2011, IPCO.

[27]  R. Weber.  On the Gittins Index for Multiarmed Bandits , 1992 .

[28]  Stefan H. Thomke,et al.  Learning by experimentation: Prototyping and testing , 2008 .

[29]  Martin L. Puterman,et al.  A note on bias optimality in controlled queueing systems , 2000, Journal of Applied Probability.

[30]  Evan L. Porteus,et al.  Optimal Timing of Reviews in Concurrent Design for Manufacturability , 1995 .

[31]  Rolf H. Möhring,et al.  Resource-constrained project scheduling: Notation, classification, models, and methods , 1999, Eur. J. Oper. Res..

[32]  Stylianos Kavadias,et al.  OPTIMAL PROJECT SEQUENCING WITH RECOURSE AT A SCARCE RESOURCE , 2009 .

[33]  T. Vishwanath.  Optimal Orderings for Parallel Project Selection , 1992 .

[34]  J. Michael Harrison,et al.  Dynamic Scheduling of a Multiclass Queue: Discount Optimality , 1975, Oper. Res..

[35]  Hayriye Ayhan,et al.  BIAS OPTIMALITY IN A QUEUE WITH ADMISSION CONTROL , 1999, Probability in the Engineering and Informational Sciences.

[36]  P. Whittle.  Multi‐Armed Bandits and the Gittins Index , 1980 .

[37]  Bruce L. Miller.  A Queueing Reward System with Several Customer Classes , 1969 .

[38]  Paul S. Adler,et al.  From project to process management: an empirically-based framework for analyzing product development time , 1995 .

[39]  J. Papastavrou,et al.  The Dynamic and Stochastic Knapsack Problem with Deadlines , 1996 .

[40]  Jan A. Van Mieghem,et al.  Price and Service Discrimination in Queueing Systems: Incentive-Compatibility of Gcμ Scheduling , 2000 .

[41]  Anton J. Kleywegt,et al.  The Dynamic and Stochastic Knapsack Problem with Random Sized Items , 2001, Oper. Res..

[42]  J. Banks,et al.  Switching Costs and the Gittins Index , 1994 .

[43]  Jr. Shaler Stidham.  Optimal control of admission to a queueing system , 1985 .

[44]  Gintaras V. Reklaitis,et al.  Simulation-optimization framework for stochastic optimization of R&D pipeline management , 2003 .

[45]  R. Weber,et al.  Optimal control of service rates in networks of queues , 1987, Advances in Applied Probability.

[46]  Kingshuk K. Sinha,et al.  The development and application of a process model for R&D project management in a high tech firm: A field study , 2011 .

[47]  Christian Artmann.  The Value of Information Updating in New Product Development , 2009 .

[48]  Christoph H. Loch,et al.  Project risk management in new product development , 2008 .

[49]  J. Pekny,et al.  Managing a Portfolio of Interdependent New Product Candidates in the Pharmaceutical Industry , 2004 .

[50]  Anton J. Kleywegt,et al.  The Dynamic and Stochastic Knapsack Problem , 1998, Oper. Res..

[51]  Saif Benjaafar,et al.  Optimal Control of an Assembly System with Multiple Stages and Multiple Demand Classes , 2010 .

[52]  J. V. Mieghem.  Dynamic Scheduling with Convex Delay Costs: The Generalized CU Rule , 1995 .