The use of Data Envelopment Analysis (DEA) to evaluate efficiency is widely discussed in production and logistics environments. The literature uses classical DEA models mainly to carry out benchmarking. This work differs from others in performing an exploratory analysis of the Charnes, Cooper, and Rhodes model (CCR) for internal benchmarking, which means evaluating a company over time. To verify the CCR model utilization, an algorithm in Python is developed to analyze: (i) the periodic applicability of the model, (ii) the results’ sensitivity to changes in the scale of indicators, and (iii) variable orientation shifts. The study reveals that CCR used for periodic performance management cannot offer an intuitive interpretation in analyzing efficiency over time since its results are restricted to a variation between 0 and 1 (0 to 100%). In addition, the model is not sensitive to changes in the scale of indicators. However, the model is very sensitive to changes in the variable orientation (number of inputs and outputs of the model). Finally, the definition of the indicators as inputs and outputs following the company’s strategy generates good efficiency results.

In practice, a controlled system consists of two distinct entities: the plant and the controller. These two entities are coupled by a feedback mechanism to give the closed-loop behavior of the controlled system. Under this structure, the controlled behavior of the system can be analyzed by simulation which is costly and time consuming. In this study, we synthesize a combined plant and controller model. We consider the practical issues and propose a method which combines the advantages of Petri nets (PN) and the supervisory controller design based on Formal Language Theory. The resultant combined model is a PN and represents the controlled closed-loop behavior of the system. This model can be used for the functional and the performance analysis of the system, verification of the controller and the control code generation.

In this paper, we present a new point of view on supervisory control of Petri nets by using routing functions instead of the traditional control places. We first show the relation between the two notions. In the second part of the paper, we illustrate the use of routing functions by showing how to compute a routing function in order to avoid starvation in general Petri nets. This control uses a continuous version of the net and a description of the evolution of the net under the form of linear algebraic equations. As for the computational part, we use algebraic polynomial geometry in the continuous case and Diophantine equations for the discrete version of the Petri net under study.

In this paper we consider two types of sessions: Guaranteed Performance (GP) and point-to-multi point Best Effort (BE). The existence of different types of users sharing the same resource raises the need for a Call Admission Control (CAC) which will enable an efficient use of the shared network. As is commonly done in todays telecommunications network, CAC is assumed to be imposed only on the GP session. We analyse the effect of bandwidth reservation for BE sessions on the performance of the BE sessions (in terms of the expected delay) and of the GP sessions (in terms of their rejection probabilities). Our analysis is based on a matrix-geometric approach.

Layered manufacturing is an additive manufacturing process where the parts are built layer by layer based on the geometry provided by a CAD system. There are several different technologies for layered manufacturing. We focus on fused deposition of ceramics. Here layers are formed by depositing material through a liquefier-head on adjacent roads in the shape of contours and/or rasters. As this technology is being pushed into new frontiers to build functional parts, the quality of the fabricated parts is becoming a major driving force for improving the technology. We present our preliminary results on an online monitoring system that can be used to detect defects. The ultimate objective is to use such a system to detect and remove the defects on any given layer prior to the next layer.

In an earlier paper, the authors (1997) developed a technique called "relative temporal analysis" to study the performance of timed Petri nets, where depending on the temporal constraints imposed on the transitions, conflict was possible between the processes modelled by the Petri net. The essence of the technique focused on the transformation of the real time trajectory of the processes into a time domain where only their relative shifts were considered. Within this domain the shifts are limited to a bounded and well defined region. Here, we extend this technique to more complex systems where structural conflict is also possible. Within a regenerative cycle of the system, each and every path should be followed according to a set of prespecified frequencies. We use the relative temporal technique to synthesize an optimal sequential controller to schedule these paths.

There are several techniques for the analysis of the dynamics of timed Petri nets. When the net is not decision free, the existing methods come short of an efficient analysis. In this paper, we present a method, we call "relative temporal analysis", to analyze the dynamics of a system with two processes and a shared resource. It is possible to find the resource utilization sequence, the waiting time period for each process and identify possible conflicts through this technique. Based on this technique we also build control charts which can be used for control purposes such as obtaining "optimal" conflict resolution schemes.

We study a Petri net model of a system composed of two processes sharing a resource. Conflicts may occur over the usage of the shared resource, thus making the system nondeterministic. Therefore, in the context of minimax algebra, it cannot be formulated as a linear system in order to compute its performance measures. However, if the sequence by which the resource is allocated to the two processes is known, we can transform the system into a decision-free net. For this system with an imposed constraint on the resource allocation frequencies, we show that the optimal allocation sequence is the most regular integer sequence satisfying that constraint. We also discuss the periodic behavior of this system under no constraints on the resource allocation frequencies.>