# Safety Integrity Level (SIL) assessment process for WSN using multi-QoS metrics

## TL;DR

This paper introduces a method to assess the safety of wireless sensor networks in industrial settings using multiple quality-of-service metrics.

## Contribution

The paper presents a novel approach to evaluate safety integrity levels in WSNs using multiple QoS metrics instead of a single metric.

## Key findings

- Multiple QoS metrics are used to assess compliance with safety integrity level targets in industrial WSNs.
- Statistical significance of QoS metric variations is evaluated using p-value to determine safety compliance.
- Non-compliant WSNs can improve safety by enhancing data communication defenses.

## Abstract

Wireless Sensor Networks are inseparable part of Industrial IoT and are deployed throughout the industrial value chain from material procurement to inventory management, storage, process monitoring and control, packaging, delivery till commissioning. In the context of industry 4.0, proliferation of WSN makes it both performance as well as safety critical due to the real time data collection, remote monitoring and machine-to-machine communication for collaboration and autonomous decision making. Deployment of WSN for safety applications thus necessitates compliance of critical QoS metrics with safety integrity levels (SIL) as demanded by the application, for example minimal delay in communication, longer network life, higher throughput, successful data detection rate, packet delivery ratio etc., While there are publications available to illustrate the assessment of safety integrity compliance of WSN, the treatments & illustrations therein are found to be limited to a single QoS metrics. However, in practice, the performance characterization of industrial WSN involves multiple QoS metrics, hence there is a need to address safety assessment use cases involving multiple QoS metrics. In this paper we are trying to bridge this gap by illustrating the safety assessment approach involving more than one QoS metrics. The approach illustrated here leverages random data simulation technique for generating QoS metrics such as the delay bound (DB) and false positive detection rate (FPDR) of a typical WSN in an industrial application environment and the statistical techniques used for consolidation of results for the decision making related to compliance with the safety integrity level. This paper is expected to serve functional safety design engineers for real time safety assessment of WSN involving more than one QoS metrics. We conclude this paper by identifying opportunities for future research in the area of safety integrity assessment of industrial WSN systems using QoS metrics.•Multiple QoS metrics are leveraged to assess compliance of WSN with safety integrity level (SIL) targets specified in industrial functional safety standard IEC 61508•The safety compliance of WSN with SIL targets is assessed considering significance of variations in all the applicable QoS metrics based on the p-value statistic.•In the event of non-compliance of QoS metrics with safety targets, data communication defences to be improved to achieve compliance.

Multiple QoS metrics are leveraged to assess compliance of WSN with safety integrity level (SIL) targets specified in industrial functional safety standard IEC 61508

The safety compliance of WSN with SIL targets is assessed considering significance of variations in all the applicable QoS metrics based on the p-value statistic.

In the event of non-compliance of QoS metrics with safety targets, data communication defences to be improved to achieve compliance.

Image, graphical abstract