Complexity in design of technical systems § Stephen C.-Y. Lu (1) a, *, Nam-Pyo Suh (1) b a University of Southern California, Los Angeles, CA 90089, USA b Korea Advanced Institute of Science and Technology (KAIST), Republic of Korea 1. Introduction As demands for function, cost, quality, and sustainability increase, the complexity of technical systems resulted from design raises rapidly. In today’s technology-driven economy, the market is full of complex technical systems that are difficult and costly to create, build, operate, maintain and recycle. Due to the high complexity, these systems are vulnerable to failures, leading to costly repairs, environmental damages, and human casualties. To achieve sustainable growths, engineers must better understand and design complex technical systems to serve the societal needs effectively with least squanders [1]. The scientific studies of complexity in design of technical systems hold the key to this major challenge. Complexity occurs in systems that have many elements with intricate dependencies among them. Due to their numerous sizes and relationships, the behaviors of complex systems are difficult to predict, even when the properties of their parts are given. As a result, complexity studies often lead to the question of probability of a system encounter a given condition once its characteristics are specified. Weaver [2] defined complexity as the ‘‘degree of difficulty’’ in predicting the properties of a system. Many approaches are developed based on this probability view to analyze and forecast the behaviors of complex systems [3]. Recently, the emergent concept has gained popularity in studying the dynamic characteristics and chaotic states of complex system [4]. For instance, cybernetics research refers to complexity as the degree of adaptability or flexibility that characterizes the system’s ability to change or emerge to survive [5]. In the Sciences of the Artificial, Simon [6] explored complexity of artificial systems with the hierarchical complexity and nearly decomposability concepts. To date, probability (i.e., uncertain future state) and emergent (i.e., time-dependent behaviors leading to uncertain state) are the most common characters used in analyzing and describing complex systems behaviors. However, there is no systematic knowledge regarding the complexity of technical systems from the design vantage. Based on the probability notion, Suh recently proposed a new Design-Centric Complexity (DCC) theory [7] which defines complexity as ‘‘the measure of uncertainty in achieving the functional requirements (FRs) of a system within their design ranges’’. The theory characterizes four types of system complexity [8]: time-independent real complexity, time-independent ima- ginary complexity, time-dependent combinatorial complexity, and time-dependent periodic complexity. It suggests that system’s complexity can be reduced by (1) minimize the number of FRs, (2) remove time-independent real complexity, (3) eliminate time- independent imaginary complexity, and (4) transform time- dependent combinatorial complexity to become time-dependent periodic complexity. Unlike those analysis-oriented approaches, the DCC theory focuses on systems that are still under creation or modification and treats complexity in the functional (not physical) domain. This treatment sheds a different light on the complexity concept than others that view complexity as an inherited and absolute character of existing systems. With the DCC theory, complexity becomes a design consideration that is to be synthesized in situ, not analyzed afterwards. In other words, complexity is not an ‘‘as-is’’ description anymore, but rather a ‘‘to- achieve’’ state, of a system of interest. This paper examines the complexity of technical systems from the design perspective. The system overall difficulty is defined as consisting of ‘‘acquired complexity’’ due to design decisions and ‘‘inborn complication’’ owing to customer needs and external constraints. Uncertainty is used as a common measure of the degrees of system complexity, complication and difficulty. The CIRP Annals - Manufacturing Technology 58 (2009) 157–160 ARTICLE INFO Keywords: Design System Complexity ABSTRACT This paper discusses the complexity of technical systems resulted from engineering design. We defined the ‘‘overall difficulty’’ of such a system as consisting of ‘‘inborn complication’’ due to customer needs and external constraints as well as ‘‘acquired complexity’’ associated with uncertainty in satisfying the functional requirements caused by design decisions. An ideal design should lead to the least difficult technical system by minimizing inborn complication and eliminating acquired complexity. To achieve this design ideality, strategies of using the Axiomatic Design theory and the Design-Centric Complexity theory are introduced to guide the creation and improvement of complex time-independent and time- dependent technical systems. ß 2009 CIRP. § This paper is submitted as a ‘‘CIRP Cooperative Work’’ between CIRP authors belonging to different institutes. The work is the result of a cooperative research performed during the first author’s sabbatical leave at the second author’s institution. * Corresponding author. Contents lists available at ScienceDirect CIRP Annals - Manufacturing Technology journal homepage: http://ees.elsevier.com/cirp/default.asp 0007-8506/$ – see front matter ß 2009 CIRP. doi:10.1016/j.cirp.2009.03.067