Examining the Impact of Environmental Variables on the Insurgent Behavior with the Aid of a GIS: Case Study of Afghanistan.

摘要

The guerilla warfare style tactics of the Afghanistan insurgents is a problem the US-UN troops face in Afghanistan. The insurgents are able to utilize the surrounding environment to gain a tactical advantage. Examples consist of suicide bombers, sniper attacks, Improvised Explosive Device (IED), and car bombs. These types of attacks utilize the surrounding environment in order to gain success. For example, the suicide bomb is more likely to have success if there is an environment full of people (e.g., busy city streets). An environment full of people makes it hard to target a threat. Sniper attacks have a greater chance of success if executed at high elevations or slopes with thick forest or high vegetation as a cover. The objective of this research is to develop a model utilizing Geographic Information System (GIS) as a framework to determine the significance of environmental variables on Afghanistan insurgency and resulting vulnerability to US-UN troops. A review of the relevant literature is conducted which helps to become familiar with the different mathematical models used to combat guerilla warfare.;Frederick Lanchester developed attrition equations for military conflict in 1916. The Deitchman (1962) attrition equation for military conflict model takes into account the area or geographical location of an insurgent when implementing ambush attacks. McCormick and Giordano's 2002 equation expands upon guerrilla warfare by adding the effects of recruitment and the population support for the insurgency and the government. These mathematical models take into account geography, terrain, population density and infrastructure when combating guerilla warfare tactics.;This research investigates the tracking and visualization of insurgent attacks based on their use of the geography, terrain, population density, and infrastructure, also referred to as Environmental Variables (EVs). The data used focuses on the EV parameters at an insurgent attack location. A GIS tracks the spatial and temporal movement of insurgents and creates a visual representation of Afghanistan. The EVs are visually represented at the insurgents' attack locations with a GIS. Logistic regression is the proposed model used to determine which EV parameters are significant to the vulnerability of US-UN troop's location. The familiarity with the EV parameters during the time of attack allows one to see which EV parameters would put US-UN troops in positions vulnerable to attacks. Logistic regression models the data from the GIS framework. The EV parameters that are significant are identified at insurgent's attack locations. A Network Analyst based GIS algorithm helps to locate the vulnerable areas that meet the specified EV criteria. The proposed model is applied to real case studies of Afghanistan insurgent attacks against US-UN troops.;The results show Afghanistan insurgents' use the EVs to increase the probability of US-UN fatalities. The years 2008, 2009 and 2010 for Afghanistan are significant with a Multiple R of 0.83966 and R Square 0.705028 with the F-test of 71%. The ANOVA for the Afghanistan database tested significant with F at 0. The regression coefficient table p-value for the EVs tested 5.7E-182 for Population Density, 1.33E-74 for Elevation, 0.006331 for Slope, and 7.55E-32 for the nearest River from the insurgent's attack location. The environmental variables' p-value tested less than 0.05. Therefore, it can be concluded that EVs are significant in creating a higher probability of a US-UN fatality. A GIS algorithm is employed, which creates a route based on the distance to maneuver around the insurgent's attack locations that contain EVs that increase US-UN fatalities. The Afghanistan GIS Database allows US-UN troops to monitor entire areas with a visual image. Then optimal routes can be identified within an Afghanistan province that would minimize the possibility of a US-UN fatality. Future works may include further refinement to the model by performing a multi-objective optimization to come up with the safest routes by simultaneously examining the impact of various EVs in a multi-objective context.;Key Words: GIS, spatial analysis, insurgents, environmental variables, Afghanistan insurgency.