Abstract
We study a variant of the Probabilistic Travelling Salesman Problem arising when retailers crowdsource last-mile deliveries to their own customers, who can refuse or accept in exchange for a reward. A planner must identify which deliveries to offer, knowing that all deliveries need fulfilment, either via crowdsourcing or using the retailer's own vehicle. We formalise the problem and position it in both the literature about crowdsourcing and among routing problems in which not all customers need a visit. We show that to evaluate the objective function of this stochastic problem for even one solution, one needs to solve an exponential number of Travelling Salesman Problems. To address this complexity, we propose Machine Learning and Monte Carlo simulation methods to approximate the objective function, and both a branch-and-bound algorithm and heuristics to reduce the number of evaluations. We show that these approaches work well on small size instances and derive managerial insights on the economic and environmental benefits of crowdsourcing to customers.

Abstract
Cardiac auscultation is the key exam to screen cardiac diseases both in developed and developing countries. A heart sound auscultation procedure can detect the presence of murmurs and point to a diagnosis, thus it is an important first-line assessment and also cost-effective tool. The design automatic recommendation systems based on heart sound auscultation can play an important role in boosting the accuracy and the pervasiveness of screening tools. One such as step, consists in detecting the fundamental heart sound states, a process known as segmentation. A faulty segmentation or a wrong estimation of the heart rate might result in an incapability of heart sound classifiers to detect abnormal waves, such as murmurs. In the process of understanding the impact of a faulty segmentation, several common heart sound segmentation errors are studied in detail, namely those where the heart rate is badly estimated and those where S1/S2 and Systolic/Diastolic states are swapped in comparison with the ground truth state sequence. From the tested algorithms, support vector machine (SVMs) and random forest (RFs) shown to be more sensitive to a wrong estimation of the heart rate (an expected drop of 6% and 8% on the overall performance, respectively) than to a swap in the state sequence of events (an expected drop of 1.9% and 4.6%, respectively).

Abstract
In this paper we describe the implementation of a secure key distribution system based on an ultra-long fiber laser with a bi-directional erbium doped fiber amplifier. The resilience of the system was tested against passive attacks from an eavesdropper, having been observed a similarity in spectrum for both secure configurations of the system.

Abstract
Orion is a NASA spacecraft being developed for human exploration of the Moon and Mars. Crew Dragon is a commercial spacecraft used to transport astronauts to and from the International Space Station (ISS). Both spacecraft are of similar design and both spacecraft perform a water landing following re-entry. This study evaluated the ability of International Institute of Astronautical Sciences (IIAS) Citizen Astronaut Candidates (CAC) to egress a spacecraft mock-up wearing a commercially available intravehicular activity (IVA) spacesuit manufactured by Final Frontier Design (FFD) (Anderson, 2014; Barker and Bellenkes; 1996; Rubio et al., 2004). This suit is similar to those worn by astronauts traveling to the ISS on board Crew Dragon. Mobility assessment revealed that most participants had sufficient ranges of motion to perform egress tasks successfully. In some instances suited participants were unable to perform selected tasks proficiently, but in these instances this often stemmed from difficulty in achieving a stable upright position in the water. Seat ingress and egress evaluation revealed no significant problems with anthropometric accommodation across participants.

Abstract
Traffic flow forecasting is an essential component of an intelligent transportation system to mitigate congestion. Recurrent neural networks, particularly gated recurrent units and long short-term memory, have been the stateof-the-art traffic flow forecasting models for the last few years. However, a more sophisticated and resilient model is necessary to effectively acquire long-range correlations in the time-series data sequence under analysis. The dominant performance of transformers by overcoming the drawbacks of recurrent neural networks in natural language processing might tackle this need and lead to successful time-series forecasting. This article presents a multi-head attention based transformer model for traffic flow forecasting with a comparative analysis between a gated recurrent unit and a long-short term memory-based model on PeMS dataset in this context. The model uses 5 heads with 5 identical layers of encoder and decoder and relies on Square Subsequent Masking techniques. The results demonstrate the promising performance of the transform-based model in predicting long-term traffic flow patterns effectively after feeding it with substantial amount of data. It also demonstrates its worthiness by increasing the mean squared errors and mean absolute percentage errors by (1.25 - 47.8)% and (32.4 - 83.8)%, respectively, concerning the current baselines.

Abstract
This study explores the resistive switching phenomena present in 4 mu m(2) amorphous Indium-Gallium-Zinc Oxide (IGZO) memristors. Despite being extensively reported in the literature, not many studies detail the mechanisms that dominate conduction on the different states of IGZO-based devices. In this article, we demonstrate that resistive switching occurs due to the modulation of the Schottky barrier present at the bottom interface of the device. Furthermore, thermionic field emission and field emission regimes are identified as the dominant conduction mechanisms at the high resistive state of the device, while the bulk-limited ohmic conduction is found at the low resistive state. Due to the high complexity associated with creating compact models of resistive switching, a data-driven model is drafted taking systematic steps. (C) 2022 Author(s).