Investigating the relationship between adverse events and infrastructure development in an active war theater using soft computing techniques

Highlights

• We investigate the relationship between adverse events and infrastructure development investments in an active war theater.

• We develop soft computing techniques (ANN, FIS, and ANFIS) for estimating the number of adverse events based on the occurrence of economic improvement projects.

• The performance of each model was investigated and compared to all other models using the calculated mean absolute percentage error (MAPE) values.

• When the model accuracy was calculated based on the MAPE for each of the models, ANN had better predictive accuracy than FIS and ANFIS models, as demonstrated by experimental results.

• The sensitivity analysis results show that the importance of economic development projects varied based on the specific regions and time period.

Abstract

The purpose of this paper is to investigate the relationship between adverse events and infrastructure development investments in an active war theater by using soft computing techniques including fuzzy inference systems (FIS), artificial neural networks (ANNs), and adaptive neuro-fuzzy inference systems (ANFIS) where the accuracy of the predictions is directly beneficial from an economic and humanistic point of view. Fourteen developmental and economic improvement projects were selected as independent variables. A total of four outputs reflecting the adverse events in terms of the number of people killed, wounded or hijacked, and the total number of adverse events has been estimated.

The results obtained from analysis and testing demonstrate that ANN, FIS, and ANFIS are useful modeling techniques for predicting the number of adverse events based on historical development or economic project data. When the model accuracy was calculated based on the mean absolute percentage error (MAPE) for each of the models, ANN had better predictive accuracy than FIS and ANFIS models, as demonstrated by experimental results. For the purpose of allocating resources and developing regions, the results can be summarized by examining the relationship between adverse events and infrastructure development in an active war theater, with emphasis on predicting the occurrence of events. We conclude that the importance of infrastructure development projects varied based on the specific regions and time period.

Introduction

Irregular warfare is defined by the U.S. Department of Defense (DoD) as “a violent struggle among state and non-state actors for legitimacy and influence over the relevant population(s).” Such warfare includes non-proportional force to convince and coerce where opposite forces are not large and effective in the region [1]. The success of irregular warfare operations depends heavily on the safety of civilian population, because the civilian population is the primary target of irregular warfare. Actors in warfare can work more effectively with a better understanding of this dynamic.

The U.S. military has made some adjustments to its force structure in response to the challenges of irregular warfare. Bhattacharjee [2] outlined the application of a program called human social culture behavior (HSCB) modeling to guide the U.S. military in achieving a better understanding of the different types of cultures encountered while operating in overseas countries [3]. The overarching aim of the HSCB modeling is to enable DoD and the U.S. Government to better organize and control the human terrain during nonconventional warfare and other missions [4]. In particular, models of human behavior could be used to predict the effects of actions intended to disrupt terrorist networks.

HSCB models are formed in order to understand the behavior and structure of organizational units at the macro level (economies, politics, socio-cultural regions) and at the micro level (terrorist networks, tribes, military units) [5]. These models are attracting much attention with regard to current and future operational requirements. HSCB models are complex systems and require computational modeling and simulation techniques to handle this complexity. Computational social scientists are currently researching how observations of human behavior might be used to develop scientifically based models of HSCB events [6]. To our best knowledge, there is currently no other study on the use of any computational methodology for representing the relationship between adverse events and infrastructure development investments in an active war theater.

Various challenges have been encountered while trying to address problems related to representing social sciences data. Numrich and Tolk [7] summarized these challenges: lack of common vocabulary, variations in modeling approaches, and data acquisition. Schmorrow et al. [8] emphasized the challenge of leveraging modeling and simulation (M&S) for HSCB. They cited the difficulty in understanding which M&S tools are actually useful and when and how best to use M&S tools within different complexity levels. All these challenges must be understood and researchers should meet specific modeling requirements before proceeding to apply various methodologies in the HSCB field.

There are modeling challenges with social data, particularly with regard to handling complexity, uncertainty, etc. Prediction is a particularly difficult task. Tools such as fuzzy inference systems (FIS), artificial neural networks (ANN), and adaptive neuro-fuzzy inference systems (ANFIS) are especially effective in handling the uncertainty and imprecision inherent in prediction modeling, and have demonstrated a high degree of accuracy in prediction applications. To demonstrate this, we’re going to use soft computing to analyze a certain characteristic we know aligns with stability and lack of violence (infrastructure and economic spending) and see if we can use it to do a really difficult but useful modeling task: predict adverse events.

In summary, this study extends the investigation into using the soft computing techniques ANN, FIS, and ANFIS for predictive modeling. In this study, we analyze real-world data, use ANN, FIS, and ANFIS to predict future behavior, and measure the accuracy of each methodology. Using data on activity in an active war theater (seven geographical regions in Afghanistan), we explore the relationship between the size and scope of infrastructure development within each region and the number of “adverse events” (people killed, wounded, or hijacked). We then use ANN, FIS and ANFIS to evaluate that relationship and predict the number of adverse events in the future.