Data aggregation has emerged as a basic concept in wireless sensor networks. By combining data and eliminating redundancy, this concept showed its effectiveness in terms of reducing resources consumption. The data aggregation problem is not new and has been widely studied in wireless sensor networks. However, most of the proposed solutions are based on a static data aggregation scheme and do not consider the network constraints evolution such as energy bandwidth, overhead and transmission delay. To dynamically optimize the tradeoff between the data aggregation process and the network constraints, a new paradigm has been introduced recently named the feedback control system. The main idea is to adapt the data aggregation degree to the environment changes and the sensor network applications. The feedback control for data aggregation remains relatively a new research area and only few works have introduced this concept. Therefore this work points out this research field by surveying the different existing protocols.

Data aggregation has emerged as a basic concept in wireless sensor networks. By combining data and eliminating redundancy, this concept showed its effectiveness in terms of reducing resources consumption. The data aggregation problem is not new and has been widely studied in wireless sensor networks. However, most of the proposed solutions are based on a static data aggregation scheme and do not consider the network constraints evolution such as energy bandwidth, overhead and transmission delay. To dynamically optimize the tradeoff between the data aggregation process and the network constraints, a new paradigm has been introduced recently named the feedback control system. The main idea is to adapt the data aggregation degree to the environment changes and the sensor network applications. The feedback control for data aggregation remains relatively a new research area and only few works have introduced this concept. Therefore this work points out this research field by surveying the different existing protocols.

Data aggregation has emerged as a basic concept in wireless sensor networks. By combining data and eliminating redundancy, this concept showed its effectiveness in terms of reducing resources consumption. The data aggregation problem is not new and has been widely studied in wireless sensor networks. However, most of the proposed solutions are based on a static data aggregation scheme and do not consider the network constraints evolution such as energy bandwidth, overhead and transmission delay. To dynamically optimize the tradeoff between the data aggregation process and the network constraints, a new paradigm has been introduced recently named the feedback control system. The main idea is to adapt the data aggregation degree to the environment changes and the sensor network applications. The feedback control for data aggregation remains relatively a new research area and only few works have introduced this concept. Therefore this work points out this research field by surveying the different existing protocols.

Data aggregation has emerged as a basic concept in wireless sensor networks. By combining data and eliminating redundancy, this concept showed its effectiveness in terms of reducing resources consumption. The data aggregation problem is not new and has been widely studied in wireless sensor networks. However, most of the proposed solutions are based on a static data aggregation scheme and do not consider the network constraints evolution such as energy bandwidth, overhead and transmission delay. To dynamically optimize the tradeoff between the data aggregation process and the network constraints, a new paradigm has been introduced recently named the feedback control system. The main idea is to adapt the data aggregation degree to the environment changes and the sensor network applications. The feedback control for data aggregation remains relatively a new research area and only few works have introduced this concept. Therefore this work points out this research field by surveying the different existing protocols.

Safety instrumented systems must be designed, built and operated to meet tolerable risk level as required regulatory agencies. This requirement is closely related to their probabilistic performance measures which are either their average probability of dangerous failure on demand (PFDavg) or their average frequency of failure (PFH: Probability of Failure per Hour). The object of this work is the SIS performances evaluation taking into account the uncertainties related to the different parameters that come into play: failure rate (λ), common cause failure proportion (β), diagnostic coverage (DC), etc. This leads to an accurate and therefore safe assessment of the safety integrity level (SIL) inherent to safety functions performed by such systems. This aim is in keeping with the requirement of the IEC 61508 standard with respect to handling uncertainty. In this paper we first explain in detail the IEC 61508 approach for handling uncertainty. Afterwards, we propose an approach that combines (i) Monte Carlo analysis (MCA) and (ii) fuzzy sets. Indeed, the first method is appropriate when representative statistical data are available (using pdf of the relating parameters), while the latter applies in the case characterized by vague and subjective information (using membership function). The proposed approach is fully supported with a suitable computer code developed under the MATLAB environment.

Safety instrumented systems must be designed, built and operated to meet tolerable risk level as required regulatory agencies. This requirement is closely related to their probabilistic performance measures which are either their average probability of dangerous failure on demand (PFDavg) or their average frequency of failure (PFH: Probability of Failure per Hour). The object of this work is the SIS performances evaluation taking into account the uncertainties related to the different parameters that come into play: failure rate (λ), common cause failure proportion (β), diagnostic coverage (DC), etc. This leads to an accurate and therefore safe assessment of the safety integrity level (SIL) inherent to safety functions performed by such systems. This aim is in keeping with the requirement of the IEC 61508 standard with respect to handling uncertainty. In this paper we first explain in detail the IEC 61508 approach for handling uncertainty. Afterwards, we propose an approach that combines (i) Monte Carlo analysis (MCA) and (ii) fuzzy sets. Indeed, the first method is appropriate when representative statistical data are available (using pdf of the relating parameters), while the latter applies in the case characterized by vague and subjective information (using membership function). The proposed approach is fully supported with a suitable computer code developed under the MATLAB environment.

Safety instrumented systems must be designed, built and operated to meet tolerable risk level as required regulatory agencies. This requirement is closely related to their probabilistic performance measures which are either their average probability of dangerous failure on demand (PFDavg) or their average frequency of failure (PFH: Probability of Failure per Hour). The object of this work is the SIS performances evaluation taking into account the uncertainties related to the different parameters that come into play: failure rate (λ), common cause failure proportion (β), diagnostic coverage (DC), etc. This leads to an accurate and therefore safe assessment of the safety integrity level (SIL) inherent to safety functions performed by such systems. This aim is in keeping with the requirement of the IEC 61508 standard with respect to handling uncertainty. In this paper we first explain in detail the IEC 61508 approach for handling uncertainty. Afterwards, we propose an approach that combines (i) Monte Carlo analysis (MCA) and (ii) fuzzy sets. Indeed, the first method is appropriate when representative statistical data are available (using pdf of the relating parameters), while the latter applies in the case characterized by vague and subjective information (using membership function). The proposed approach is fully supported with a suitable computer code developed under the MATLAB environment.