Summary form only: Abstracts of articles presented in this issue of the publication.

Summary form only: Abstracts of articles presented in this issue of the publication.

Summary form only: Abstracts of articles presented in this issue of the publication.

Summary form only: Abstracts of articles presented in this issue of the publication.

“Scanning the Issue.“

Summary form only: Abstract of article "Scanning the Issue."

Summary form only: Abstracts of articles presented in this issue of the publication.

Summary form only: Abstracts of articles presented in this issue of the publication.

Summary form only: Abstracts of articles presented in this issue of the publication.

Summary form only: Abstracts of articles presented in this issue of the publication.

Summary form only: Abstracts of articles presented in this issue of the publication.

The presentation deals with the possibility of defining control actions for vehicle platoons in a road traffic network not only devoted to an effective behaviour of platoon themselves, but also considering them as possible actuators to mitigate traffic congestion. It is known that the formation of platoons presents relevant advantages for the involved vehicles, first of all in terms of fuel consumption. Besides this, in this presentation, the concept of platoon is further developed by investigating the possibility of exploiting the presence of platoons also for relieving congestion for the whole traffic stream. In this sense, each platoon can be considered as a moving bottleneck that affects the surrounding traffic conditions depending on its size (i.e., the number of vehicles composing it), the position of vehicles, and its speed. By appropriately acting on these aspects, each platoon can be adopted as an actuator of traffic management strategies. Several control schemes have been proposed in the literature, from simple feedback regulators to model predictive control schemes, both in centralized architectures and in decentralized ones. The different schemes are presented and their effectiveness, analyzed by means of simulations, is discussed. The main open issues, related to modelling aspects, to information management requirements and to the practical applicability of the ideas behind the research work, are also introduced in order to evaluate opportunities and challenges.

It covers the fundamental analysis of sensors and sensor fusion techniques for autonomous vehicles. It focuses on the current challenges in sensor fusion and recent achievements of sensor fusion in the light of technical evaluations. It explains the different sensors on the base of the advantages and disadvantages of AVs. It includes deep learning and traditional approaches for surrounding object detection, object recognition, and other vehicle movement detection using different sensors. It also provides cutting-edge assessment matrices for comparing quantitative and qualitative characteristics of various techniques.

A Survey of Machine Learning-Based Ride-Hailing Planning

The authors provide a meta-review across nearly 100 extant surveys on unmanned aerial vehicles (UAVs) applications and overview papers, extract their key messages, and investigate the extent of being complementary. Based on the findings of their survey analysis, they developed the so-called AERIAL framework, which aggregates the major challenges on the way to a successful application of UAVs for logistics, mobility, and monitoring. The framework covers the following aspects: 1) acceptance and heterogeneities; 2) economy and demand; 3) reliability and risks; 4) infrastructure and optimization; 5) automation and control; and 6) liability and standardization. They believe that their AERIAL framework and meta-review contribute towards a clearer understanding of the scientific UAV landscape challenges and the identification of potential directions for future research studies.

Toward Effective Traffic Sign Detection via Two-Stage Fusion Neural Networks

Summary form only: Abstracts of articles that will be presented in future issues of this publication.

Summary form only: Abstracts of articles presented in this issue of the publication.

Summary form only: Abstracts of articles presented in this issue of the publication.

Summary form only: Abstracts of articles presented in this issue of the publication.

In the European Rail Traffic Management System (ERTMS), the Route Control Centre System (RCCS) supervises the distance between consecutive trains and generates movement authorities, i.e. the permission for a train to move to a specific location within the constraints of the infrastructure and with supervision of speed. In this work, a control model aimed at determining the speed and position of a train platoon within a sector of the rail network is presented. The central controller, i.e. the RCCS, receives information about the current position and speed of trains and it sends them decisions about optimal corrective actions for each train. Priorities of trains are handled to respect the planned timetable, taking into account the train dynamics, limitations in divergences of positions, speeds, and tractive effort, as well as minimum distances between consecutive trains. The control approach is based on a quite innovative linear quadratic regulator allowing the definition of stochastic constraints. The validation of the model is based on data collected from a RCCS for a Section of the high-speed Paris-London line.

The efficiency of freeway systems is often hindered by recurrent and non-recurrent congestion. The occurrence of these events is due not only to the high number of vehicles trying to use a shared infrastructure, but also by phenomena that temporarily deteriorate the capacity of the infrastructure itself. Among them, road accidents are considered as one of the primary causes of non-recurrent congestion. At the same time, several studies also identify traffic breakdowns as events leading to vehicle crashes. A specific goal of this research is to develop a global safety index that quantifies the expected number of crashes as a function of the current traffic state in the freeway system. Moreover, on the basis of this new index and the performance indicator generally adopted to evaluate the traffic delay, a coordinated ramp metering scheme is proposed jointly considering the reduction of travel times for the drivers and the improvement of safety in the freeway system. The control strategy is sought by defining a nonlinear optimal control problem with constrained control variables, solved by applying a specific gradient-based algorithm. The simulation analysis investigates the multi-objective nature of the problem assessing whether and to what extent the two components of the cost criterion are conflicting objectives.

This paper proposes an optimization approach for sizing port rail networks and planning railway shunting operations by adopting a discrete-time model of the overall system. First, a mixed-integer linear mathematical programming problem is defined in order to optimize shunting operations to be performed on the considered network by satisfying certain arrivals and departures of import and export flows. Moreover, the proposed procedure can be used to evaluate the capacity of a port rail network, in terms of maximum number of trains that can be managed over a certain time horizon, and to carry out what-if analyses aimed at testing different scenarios. The effectiveness of the proposed approach is shown by applying the optimization problem to a real case study referred to the port rail network of La Spezia Container Terminal located in Northern Italy. A computational analysis realized by varying the dimension and complexity of the problem instances is also reported in this paper to discuss the computational performance of the proposed model.

This paper deals with a distribution system with one origin and one destination in which the shipments are realized with a limited fleet of capacitated vehicles. An optimization problem is defined to find the optimal shipment policies in order to satisfy the external demand, by minimizing holding costs of the customer and transportation costs. Note to Practitioners - A distribution system with one origin and one destination characterized by a Vendor-Managed Inventory policy is considered. In such a system the supplier plans the deliveries to the customer in order to minimize the customer's inventory and the transportation costs. Besides the definition of some solution methods based on mathematical programming techniques, an algorithmic approach for solving the considered optimization problem is also derived in this paper. Even though this approach is suboptimal, it provides very effective solutions and its major strength stands in the possibility of easily implementing it without the use of mathematical programming solvers.

This paper proposes a planning approach to optimize railway operations in seaport terminals by adopting a queue-based discrete-time model of the considered system. First, a mixed-integer linear mathematical programming problem is defined in order to optimize the timing of import trains and the use of the handling resources devoted to rail port operations. Second, in order to deal with unexpected situations or uncertainty in estimating some data necessary to the planning, an event-triggered receding-horizon planning approach is proposed, in which the finite horizon optimization problem is solved whenever a critical event happens or the real values of some problem data significantly differ from the predicted ones. Both these planning approaches are tested on data referred to a real terminal and deeply discussed in this paper.

Objective of this paper is to define an innovative control scheme to reduce congestion in freeway traffic systems via ramp metering. Such a control scheme is event-based and is composed of two control levels. At the lower level, Model Predictive Control is applied: the cell transmission model, reformulated in the so-called mixed-logical dynamical form, is adopted as internal model for the prediction and the quadratic deviations of the system variables from specific set-points are used as objective function in the finite horizon optimal control problem. The higher level of the control scheme consists of a discrete-event supervisor capable of distinguishing between regular and specific traffic conditions, adjusting the control law accordingly. More precisely, on the occurrence of classified events, the discrete-event supervisor determines the most suitable control action by making the controller switch among the elements of a set of model predictive controllers based on different internal models and cost functions.

A specific class of production systems is considered in this paper with the aim of minimizing inventory and production costs, over a finite horizon. In such a class of systems, all processes (arrival and departure of parts, production, demand to be satisfied) are modelled as piece-wise constant functions that change their values at asynchronous time instants. The decision variables are the constant rate at which raw parts enter the system and the service rate at the various time instants. In the paper, both the case of completely known demand and that of incomplete demand is taken into account. In both cases, a solution algorithm is provided in order to determine the optimal (or sub-optimal, in case of incomplete demand) solution of the optimization problem.

In this paper, supply chain nodes are considered, with the aim of optimizing the production of parts belonging to different classes (the case of two classes is here taken into account). The production node is modelled by a discrete-event model, being the system state, consisting of the continuous level of input and output inventories, affected by the asynchronous processes relevant to part arrivals and product departures. Moreover, also the resource capacity of the node is represented by a continuous variable (to be shared among the two classes). In the paper, two optimization problems are proposed: the former is only stated due to its intrinsic complexity, whereas the latter (a simplified version of the former) is stated and solved. In the latter problem, an optimization is performed each time an event occurs, in order to determine which the next event is, when the next event will occur, how many raw parts/products will arrive/depart, and which portions of production capacity are assigned to the two classes.

The optimal control of a single machine processing a certain number of jobs and modelled as a discrete-event dynamic system is considered. The sequence of jobs is fixed, whereas their timing and sizes represent the control variables of the system. The objective function to be optimised is a weighted sum of the inventory cost and of the quadratic deviations of job lot-sizes and completion times from those specified by the production demand. The problem is formalised as a multistage optimal control problem and solved by a procedure making use of dynamic programming techniques; in this way, a proper real time control scheme is obtained.

This paper formalizes a unique decisional framework to state different problems relevant to automated manufacturing systems, such as production planning, material requirements planning, and online management. Since the manufacturing systems are modelled as discrete-event systems, the (timed) Petri net formalism can be adopted to represent their dynamics. Moreover, in order to make the decisional problems addressed computationally tractable, it is assumed that the manufacturing systems can be represented by means of a particular class of timed Petri nets-the timed event graphs. Two particular optimization problems are discussed in the paper: the first one is relevant to the minimization of production and inventory costs, while satisfying a pre-specified production demand; and the second one refers to an online management problem in which the deviations between the actual system behaviour and a pre-defined nominal behaviour are sought to be minimized.

A model of manufacturing systems is considered in which several classes of products are repeatedly manufactured, according to the requirements specified by customer orders. The problem considered is the development of control strategies capable of reacting to deviations of the system behaviour from a nominal behaviour defined solving a suitable problem at the scheduling level. The degrees of freedom allowed for the control problem are: the execution times of some elementary operations, the lot-sizes, and the variables expressing the timing of the production process. It is shown that, under suitable hypotheses about the nominal behaviour, the system can be represented by means of a particular class of timed Petri nets. Moreover, the structure of the control problem is that of a continuous variable problem with quadratic cost and linear (or sometimes quadratic) constraints.

A novel hybrid model for transportation systems is proposed, in which the functioning is represented via a continuous-time model whenever no changes in the traffic flow occur, whereas the changes in the system behaviour are modelled through a discrete event system coupled with the continuous-time system. Via the discrete-event part of the hybrid model, its continuous-time part is adapted according to the variation occurred in the system behaviour. The hybrid modelling methodology proposed is based on the use of a special class of Petri nets, the hybrid stochastic Petri nets (HSPN). These nets are composed of some discrete parts, and some continuous parts, so that a hybrid Petri net can be viewed as a combination of an usual (discrete) Petri net and a continuous Petri net, the 'fluid' version of usual timed Petri nets. The discrete part of the HSPN 'regulates' the continuous PN, which models the functioning of the transportation system. This results in the model reacting to any change to the current system behaviour by adapting the traffic parameters to the new state.

A model of a cyclic manufacturing system representable as a deterministic timed event graph is considered in this paper. It is possible to apply analytical techniques for performance evaluation of such systems. On this basis, a two-level optimization problem is considered. The higher level refers to the maximization of the system productivity with respect to the assignment of operations to machines, the lot sizes, and the service sequences at the various machines, assuming that unlimited work-in-progress and buffers of finite dimension are available. The lower level deals with the minimization of an aggregate cost taking into account the work-in-progress and the buffer dimensions.