We provide an overview and comparison of predictive capabilities of several methods for ranking association football teams. The main benchmark used is the official FIFA ranking for national teams. The ranking points of teams are turned into predictions that are next evaluated based on their accuracy. This enables us to determine which ranking method is more accurate. The best performing algorithm is a version of the famous Elo rating system that originates from chess player ratings, but several other methods (and method versions) provide better predictive performance than the official ranking method. Being able to predict match outcomes better than the official method might have implications for, e.g., a team's strategy to schedule friendly games.

We address the problem of dynamic ambulance repositioning, in which the goal is to minimize the expected fraction of late arrivals. The decisions on how to redeploy the vehicles have to be made in real time, and may take into account the status of all other vehicles and accidents. This is generally considered a difficult problem, especially in urban areas, and exact solution methods quickly become intractable when the number of vehicles grows. Therefore, there is a need for a scalable algorithm that performs well in practice.We propose a polynomial-time heuristic that distinguishes itself by requiring neither assumptions on the region nor extensive state information. We evaluate its performance in a simulation model of emergency medical services (EMS) operations. We compare the performance of our repositioning method to so-called static solutions: a classical scenario in which an idle vehicle is always sent to its predefined base location. We show that the heuristic performs better than the optimal static solution for a tractable problem instance. Moreover, we perform a realistic urban case study in which we show that the performance of our heuristic is a 16.8% relative improvement on a benchmark static solution. The studied problem instances show that our algorithm fulfils the need for real-time, simple redeployment policies that significantly outperform static policies.

In this paper we study the scheduling of jobs with a constraint on the average waiting time in the presence of background jobs. The objective is to schedule to s servers such that the throughput of the background traffic is maximized while satisfying the response time constraint on the foreground traffic. A typical application of this model is call blending in call centers. Here the servers are called agents, the foreground traffic are the incoming calls and the background traffic are the outgoing calls. The arrivals are determined by a Poisson process and the service times of the jobs are independent exponentially distributed. We consider both the situation where service requirements by both types of jobs are equal and unequal. The first situation is solved to optimality, for the second situation we find the best policy within a certain class of policies. Optimal schedules always keep part of the service capacity free for arriving foreground jobs.

No abstract