93 Journal of Environmental Informatics ISEIS Journal of Environmental Informatics 21(2) 93-101 (2013) www.iseis.org/jei Interval Binary Programming for Noise Control within an Urban Environment W. Huang 1,* , L. M. Dai 2 , B.W. Baetz 1 , M. F. Cao 2 , and S. Razavi 1 1 Department of Civil Engineering, McMaster University, Hamilton, ON L8P 2R3, Canada 2 Faculty of Engineering and Applied Science, University of Regina, Regina, Saskathewan S4S 0A2, Canada Received 12 April 2012; revised 10 July 2012; accepted 12 September 2012; published online 27 June 2013 ABSTRACT. This paper introduces an interval binary programming (IBP) method to the selection of control measures for noise reduction under uncertainty, by incorporating the concepts of interval numbers and interval mathematical programming into a binary programming optimization framework. As an extension of the binary programming method, IBP can explicitly address complexities and uncertainties in a noise control system. Parameters in the IBP model can be expressed as intervals, and uncertainties are effectively incorporated within the model solution process. The modelling approach is applied to a representative control measure selection problem for noise reduction in an urban environment. Results of the application indicate that useful solutions for noise control practices can be generated. A number of decision alternatives have been obtained and analyzed under different acceptable noise levels for two communities, and they reflect complex tradeoffs between environmental and economic considerations. Keywords: interval binary programming, uncertainty Analysis, noise control, system optimization 1. Introduction The problems associated with acoustic noise continues to be a major challenge for urban communities throughout the world due industrialisation and urbanisation. The noise gene- rated from plants and factories can pose significant threats to the health of workers and residents in nearby communities (King and Davis, 2003). Noise-induced hearing loss is one of the most common occupational diseases and the second most self-reported occupational illness or injury in the United States (Murray-Johnson et al., 2004). According to the National Ins- titute for Occupational Safety and Health (NIOSH), appro- ximately 30 million U.S. workers are currently exposed to noise hazards on the job and an additional 9 million U.S. wor- kers are at-risk for developing hearing loss (NIOSH, 1998). Long-term exposure to excessive noise levels is recognized as the major cause of hearing loss. Since hearing loss is difficult to cure, appropriate engineering controls are strongly recom- mended to minimize noise and diminish the noise effect on workers and nearby residences. However, engineering controls differ in cost and noise reduction capability; more effective noise control measures usually require greater investment, whi- le less effective measures may have lower costs. Therefore, optimization models are desired for helping decision makers make tradeoffs between system cost and noise control effi- ciency. * Corresponding author. Tel.: +1 306 5913508; fax: +1 905 5299688. E-mail address: wendywanzhihuang@gmail.com (W. Huang). ISSN: 1726-2135 print/1684-8799 online © 2013 ISEIS All rights reserved. doi:10.3808/jei.201300236 In the past decades, significant efforts have been made in developing optimization models for noise control systems. For example, Yeh et al. (2004) developed an optimization mo- del for noise reduction in a multiple noise system by using a genetic algorithm. Asawarungsaengkul and Nanthavanij (2006) proposed six optimization models for identifying the optimal noise hazard control strategy, including two models for engi- neering controls, two for job rotation and two for the use of hearing protection devices. They then applied an algorithmic approach to the selection of engineering controls for optimal noise redution (Asawarungsaengkul and Nanthavanij, 2007). Zachary et al. (2010) developed a multi-impact optimization model to reduce aviation noise and emissions at Luxem- bourg’s Findel Airport. Prats et al. (2011) proposed a multi- objective optimization model for designing aircraft noise aba- tement strategies. Also, there are a number of other optimi- zation models for identifying optimal noise control strategies (Waly and Sarker, 1998; King and Davis, 2003; Mun and Cho, 2009; Tokmechi, 2011). In a practical noise control system, many parameters such as noise-reduction effects of different control measures, the unit cost of each measure, and acceptable noise levels for receptors may have some levels of uncertainty. However, pre- vious optimization models are deterministic and only deal with parameters with crisp values. Therefore, in this paper, an interval binary programming method will be developed and applied to a representative noise control system for selecting optimal noise reduction measures. Interval solutions for bi- nary variables will be analyzed and interpreted to provide useful decision alternatives for controlling noise from different sources and thus demonstrate the potential applicability of the developed method.