The Interval Linear Programming A Revisit英文文献.pdfVIP

The Interval Linear Programming: A Revisit F. Zhou 1, H.C. Guo 1, *, K. Huang 1, G.H. Huang 2 1 College of Environmental Sciences, Peking University, Beijing, 100871, P.R. China. 2 Faculty of Engineering, University of Regina, Regina, SK, S4S 0A2, Canada Abstract. Interval linear programming (ILP) was developed by Huang and Moore (1993) and was widely applied in environmental and resources management. However, the feasibility of optimal solutions x ± [x − ,x + ] for j opt jopt j opt ∀j affect directly in generating several decision alternatives, thus a modified interval linear programming (MILP) model is developed to assure its solution space could be absolutely feasible, and its solution algorithm is proposed to incorporates the associated extra constraints into upper-/ lower- bounds submodels. Moreover, the proofs for determining A± and B± in corresponding constraints are revised and refined in this study. The results of numeric example further indicated the feasibility and effectiveness of the MILP model. Keywords: Linear programming; optimization; uncertainty; feasibility 1. Introduction Many of the input parameters in real-world problems exhibit some level of uncertainty due to the scarcity of data (Dantzig, 1955; Huang et al., 1995; Chinneck and Ramadan, 2000). This includes objective function costs (C), constraint coefficients (A), and right-hand sides (B). Thus, it is important find ways of using LP methods with intrinsic uncertainties in probabilistic, possibilistic, and/or interval formats (Maqsood et al., 2005). The methods developed to do this could be grou