Machine Learning for Warpage Prediction of Fused Deposition Modelling Processed Parts

This paper provides a methodology for the application of a machine learning-based framework for fused deposition modelling manufacturing. The approach was developed to take into account the influence of the material, the part geometry, the process parameters on the maximum part warpage defined by the user. The results showed the effectiveness of machine learning for both classification and regression purposes so that the printability of the part is firstly provided, based on the selected warpage threshold, and secondly, the part warpage can be predicted within the problem design space variables, i.e. part material, part height, part length, and layer thickness. The limitations of the use of the analytic equation as a data-points generator are widely discussed, along with the future research based on the obtained preliminary results. In conclusion, the described methodology represents a concrete step towards a first-time-right strategy in the field of manufacturing processes.

Solving high-dimensional optimal stopping problems using deep learning

Nowadays many financial derivatives, such as American or Bermudan options, are of early exercise type. Often the pricing of early exercise options gives rise to high-dimensional optimal stopping problems, since the dimension corresponds to the number of underlying assets. High-dimensional optimal stopping problems are, however, notoriously difficult to solve due to the well-known curse of dimensionality. In this work, we propose an algorithm for solving such problems, which is based on deep learning and computes, in the context of early exercise option pricing, both approximations of an optimal exercise strategy and the price of the considered option. The proposed algorithm can also be applied to optimal stopping problems that arise in other areas where the underlying stochastic process can be efficiently simulated. We present numerical results for a large number of example problems, which include the pricing of many high-dimensional American and Bermudan options, such as Bermudan max-call options in up to 5000 dimensions. Most of the obtained results are compared to reference values computed by exploiting the specific problem design or, where available, to reference values from the literature. These numerical results suggest that the proposed algorithm is highly effective in the case of many underlyings, in terms of both accuracy and speed.

THE USE OF A PIPE CUTTER AS AN INSTRUMENT OF CRIMINAL OFFENSE AT THE TIME OF THEFT OF CABLE AND WIRING PRODUCTS: FEATURES OF IDENTIFICATION RESEARCH

The article deals with a widespread modern device for plumbing work - a pipe cutter for polypropylene pipes, as an adapted instrument of crime in theft of cable and wire products, primarily a telephone cable. The main attention is focused on the following aspects of the problem: design features of a pipe cutter with a ratchet mechanism, in the context of solving identification issues; systematization of traces that the pipe cutter leaves on the surfaces of the constituent elements of the telephone cable; problematic in the study of cut traces on a telephone cable. The structure of the pipe cutter is considered in detail, with an indication of its main parts. In accordance with the design features, the type of the considered tool is determined, relative to the existing modern forensic classification. The features are generalized and systematized that leave the components of the pipe cutter, the complex of which allows you to identify tools of this type in research without a comparative sample. The main stages of the mechanism of trace formation are modeled, understanding of which makes it possible to trace the areas of the contacting surfaces of the tool. A visual diagram of the phased movement of the pipe cutter blade has been drawn up, which illustrates the mechanism of the formation of certain general and particular features in the cut marks, including simplifying the solution of identification problems. The article considers the possibility of compiling an information retrieval model of a crime instrument, as well as the effectiveness of its use in operational search activities and the investigation of crimes related to theft of a telephone cable. The main problematic points identified in practice, in comparative traceological studies of the cut traces left by the pipe cutter, are generalized and explained, among which the most common are a relatively short identification period, the instability of the trace formation mechanism and the limited surface area of the telephone cable elements, on which informative traces remain.