No abstract

No abstract

Unmanned operations and automation in modern industry create complex everyday problems, which require algorithmic thinking and creativity. Development of risk assessment methods is critical for the future of this business segment. To provide decision support for the management of an autonomous emission control boat, we begin by proposing a k‐Nearest‐Neighbours (k‐NN)‐based trajectory prediction method. This is employed in a bi‐objective routing problem of finding a Hamiltonian circuit in a dynamic network defined by predicted locations of ships over time. The objectives are maximizing the number of measurement tasks to be done and minimizing the corresponding total travel distance of the emission control boat. To evaluate the impact of trajectory prediction uncertainty on Pareto‐optimal itineraries, we propose a risk measure in a mean‐risk framework. The risk is defined based on an expected shortfall when implementation of an efficient itinerary under the predicted trajectories needs rescheduling based on realized trajectories. The risk measure helps the decision maker to evaluate choice alternatives among efficient itineraries under predicted trajectories and to make a balanced risk‐adjusted decision. We show how historical data is employed in integer linear programming for the estimation of such risk measure. Empirical results demonstrate such estimation.

Several objective functions have been proposed in the literature to adjust the input parameters of a node in constructive networks. Furthermore, many researchers have focused on the universal approximation capability of the network based on the existing objective functions. In this brief, we use a correntropy measure based on the sigmoid kernel in the objective function to adjust the input parameters of a newly added node in a cascade network. The proposed network is shown to be capable of approximating any continuous nonlinear mapping with probability one in a compact input sample space. Thus, the convergence is guaranteed. The performance of our method was compared with that of eight different objective functions, as well as with an existing one hidden layer feedforward network on several real regression data sets with and without impulsive noise. The experimental results indicate the benefits of using a correntropy measure in reducing the root mean square error and increasing the robustness to noise.

No abstract