Piecewise linear lower and upper bounds for the standard normal first order loss function

Abstract

The first order loss function and its complementary function are extensively used in practical settings. When the random variable of interest is normally distributed, the first order loss function can be easily expressed in terms of the standard normal cumulative distribution and probability density function. However, the standard normal cumulative distribution does not admit a closed form solution and cannot be easily linearised. Several works in the literature discuss approximations for either the standard normal cumulative distribution or the first order loss function and their inverse. However, a comprehensive study on piecewise linear upper and lower bounds for the first order loss function is still missing. In this work, we initially summarise a number of distribution independent results for the first order loss function and its complementary function. We then extend this discussion by focusing first on random variable featuring a symmetric distribution, and then on normally distributed random variables. For the latter, we develop effective piecewise linear upper and lower bounds that can be immediately embedded in MILP models. These linearisations rely on constant parameters that are independent of the mean and standard deviation of the normal distribution of interest. We finally discuss how to compute optimal linearisation parameters that minimise the maximum approximation error. keywords: first order loss function; complementary first order loss function; piecewise linear approximation; minimax; Jensen's; Edmundson-Madansky.

The stochastic lot sizing problem with piecewise linear concave ordering costs

“…We refer the interested reader to Rossi et al (2014) for a detailed analysis of piecewise linear approximations of the standard loss function.…”

The Dynamic Bowser Routing Problem

“…. + f a t ; these expressions are easily generalised to the case in which random variables are continuous [55].…”

“…Constraints (20) and (21) can be easily implemented via piecewise linearization techniques presented in [55,54] and via the piecewise command in IBM ILOG OPL [38]. Constraint (22) can be implemented using the IBM ILOG OPL maxl command.…”

Computing non-stationary (s, S) policies using mixed integer linear programming

“…We assume that demand d t in each period t is independent and normally distributed with meand t and coefficient of variation c v ∈ {0.1, 0.2, 0.3}; note that σ t = c vdt . Since we operate under the assumption of normality, our models can be readily linearised by using the piecewise linearisation parameters available in Rossi et al (2014). However, the reader should note that our proposed modeling strategy is distribution independent, see Rossi et al (2015).…”