Plenary Speaker
Quicklinks to the plenary speakers:
Shane Henderson
Ton de Kok
Kai Furmans
Talk:
Real-Time Control of Ambulance Fleets, and Simulation Optimization using High-Performance Computing
In the first part of the talk I will discuss ambulance redeployment, in which an ambulance fleet is controlled in real-time to attempt to ensure short response times to calls. I'll focus on the use of simulation optimization to tune approximate-dynamic programs that yield highly effective policies, along with a coupling approach to compute a bound on the optimality gap. This work has motivated us to develop simulation-optimization algorithms that exploit parallel computing capabilities. In the second part of the talk, I'll discuss our work in developing "ranking and selection" algorithms for high-performance computing environments, and show results for runs using up to 1000 cores.
Talk:
Managing complexity and uncertainty with Newsvendor equations
Real-world problems in operations management concern multiple items to be sold, processed and sourced and multiple resources in type and quantity to be used. This complexity has been successfully dealt with in the context of deterministic problem formulations due to advances in generic optimization tools, such as CPLEX. Unfortunately, real-world problems show uncertainty in demand and processes, whereby objective functions become nonlineair and in complex situations even expressions for objective functions and constraints may be hard to derive. Despite this statement it seems that for generic multi-item multi-echelon problems optimal policies within well-known classes of policies, e.g. base stock policies, satisfy so-called Newsvendor equations. These equations are characterized by target customer service fractiles that can be computed directly from holding and penalty cost parameters, where holding costs are charged for inventories and penalty costs are charged for backlogs or lateness. The form of these equations enables efficient computation of optimal policies, despite the non-linear multi-dimensional objective functions. In this presentation we provide insight into the nature of the problems that leads to the emergence of Newsvendor equations emerge. We also discuss the managerial insights that follow directly from the target fractiles.
Talk:
Stochastic Models in Lean Management – Applications and Research Opportunities
Lean Management was first viewed as a collection of tools like Kanban, internal and external Milkruns and kaizen. Later it was understood, that at the core of it, it is a successfull approach in leadership and management.
However, when analyzing the tools, it can be shown by using stochastic models, how the individual tools support the overall goals of lean management. Using a few examples, it is shown, how stochastic models can be used, to evaluate the impact of lean methodologies on the performance of an operation.
However, there are still some areas, where lean management does not have answers yet. thus giving opportunities for further research. Some examples will be given.
