Thinking strategically about thinking strategically: the computational structure and dynamics of managerial problem selection and formulation
提出一个管理者问题表述的新模型，基于计算复杂性将问题分为P型和NP型，解释管理者为何偏好解决简单问题，并揭示常见决策偏差和启发式的成因。
Abstract A new model of managerial problem formulation is introduced and developed to answer the question: ‘What kinds of problems do strategic managers engage in solving and why?’ The article proposes that a key decision metric for choosing among alternative problem statements is the computational complexity of the solution algorithm of alternative statements. Managerial problem statements are grouped into two classes on the basis of their computational complexity: P‐type problems (canonically easy ones) and NP‐type problems (hard ones). The new model of managerial cognitive choice posits that managers prefer to engage with and solve P‐type problems over solving NP‐type problems. The model explains common patterns of managerial reasoning and decision making, including many documented ‘biases’ and simplifying heuristics, and points the way to new effects and novel empirical investigations of problem solving‐oriented thinking in strategic management and types of generic strategies, driven by predictions about the kinds of market‐ and industry‐level changes that managers will or will not respond to. Copyright © 2009 John Wiley & Sons, Ltd.