Article Preview
Top1. Introduction
The nature of warfare in the 21st century has transitioned to an unconventional approach. The accumulation and deployment of CBRN (chemical, biological, radiological, or nuclear) weapon systems is a major threat owing to their devastating potential. A comparison on the effects of chemical and biological agents on protected and un-protected troops as compared to nuclear weapons shows that the development and deployment of chemical and biological agents requires lower levels of technology and cost, thus making them easily accessible. Chemical warfare (CW) employs the properties of chemical agents as weapons. Chemical weapons were first used conventionally in World War 1. Currently, customary international humanitarian law prohibits the use of CW agents but reports of the use of chemical weapons have been received from conflicts across the world, including Syria (T. Schneider & T. Lutkedend, 2019).
MOPP is gear used by military personnel during a CBRN (chemical, biological, radiological, or nuclear) strike. MOPP protection levels range from MOPP Ready to MOPP Level 4, each level corresponds to an increased level of protection. The level is dictated by commanders in-theatre. MOPP open procedures are initiated to prolong soldier endurance in MOPP level 3 or level 4. MOPP open may also depend on local weather conditions. Sustained wearing of MOPP overgarment in level 3 or 4 can place great physical and psychological stress on the soldier. In-theatre commanders must ensure that they determine the MOPP open time correctly. If MOPP open procedures are carried out too late, soldiers stand the risk of physical and mental exhaustion. If MOPP open procedures are carried out too early, soldiers stand the risk of exposure to chemical agents.
A hierarchal adaptive neuro-fuzzy inference system (HANFIS) to predict wait times before executing MOPP open procedures is proposed. Fuzzy logic provides a suitable modelling approach to transition between complex human decision-making and logical computer calculations. The fuzzy rules and their relationships are improved using artificial neural networks. ANFIS therefore is an adaptive approach to creating fuzzy inference systems that can effectively emulate human decision making in the presence of noise or imprecise data (Sreeharsha, P. & Devanaboyina, V., 2018). As per our knowledge, no research paper explores the application of soft computing techniques to MOPP procedures in CBRN environments. Typically, in-theatre commanders depend on field manuals and standard operating procedures (SOP) to determine MOPP procedures and MOPP open times. These SOPs may not be effective for all weather conditions. The HANFIS model proposed in this paper can easily be adapted to specific data and requirements of the user. The linguistic nature of fuzzy decision support systems, combined with the computational efficiency and improved interpretability of hierarchal fuzzy systems explored in this paper will allow in-theatre commanders to:
- •
Customize MOPP analysis and hazard duration predictions according to their own criteria and situation.
- •
Define MOPP analysis ruleset for operations, depending upon the CW agents in the enemy’s arsenal, as well as local weather conditions.
- •
Increase and improve existing MOPP analysis processes with new fuzzy rules.
The applications of the proposed model are certainly not limited to MOPP open time prediction. Classification and prediction of physical exertion on soldiers due to sustained wearing of MOPP over garments, decision support system for chemical vulnerability assessment of brigade and battalion level targets can be carried out using this approach and will be carried out in future works. These systems can be combined to create an effective decision support system for battlefield commanders that assist them in all aspects of force protection in CBRN environments. This paper is divided in 5 sections for better assimilation. Section 1 is the introduction; Section 2 covers chemical warfare agents and section 3 covers MOPP equipment. Section 4 covers an overview of ANFIS. Section 5 covers the proposed model and its usage. Section 6 concludes the paper.