Abstract
The ODD Protocol has become a standard for documenting and describing agent based models. The protocol is organized around three main elements, from which the ODD acronym originates: Overview, Design concepts, and Details. This chapter is organized around these primary elements and further broken down into seven sub-elements to provide a clear purpose and understanding of the model simulation. The sub-elements are: Purpose, State Variables and Scales, Process Overview and Scheduling, Design Concepts, Initialization, Input, and Sub-models. The model presented is a proto-agent behavioral model and is utilized in an agent based modeling simulation to help identify possible emergent behavioral outcomes of the populations in the area of interest. By varying the rules governing the interactions of the multinational and incumbent government proto-agents, different strategies can be identified for increasing the effectiveness of those proto-agents and the utilization of resources.
TopThe Odd Protocol
Verification testing of this model was conducted by observing the outputs of the model at the extremes of the simulation to ensure that the results moved in predicted directions. Validation of the model involved comparisons of the model outcomes to a series of case studies involving counterinsurgency operations and the delivery of development aid. The case studies utilized included:
- •
Police in the lead with military support or vice versa (Sepp, 2004).
- •
The timing of development aid delivery vis-à-vis the stage of the counterinsurgency campaign – either early or late (Barlow, 2010).
- •
The integration of local population into security forces – either high or low (Barton, 2010; Megahan, 2010; Sepp, 2004).
- •
The level of local population inclusion in development aid delivery – either high or low (Brinkerhoff, 2010; Guttieri, 2010; Pimbo, 2010).
- •
The level of local institutional development - either high or low (Brinkerhoff 2010; Pandya, 2010; Sepp, 2004; von Hippel, 2010).
- •
The level of security from violence - either high or low (Guttieri, 2010; Sepp, 2004).
- •
The level of local population cooperation with counterinsurgency forces vis-à-vis the insurgents - either high or low (Galula, 1964).
- •
The level of coordination with government agencies – either high or low (Curry, 2010; Szayna, et al., 2009).
- •
The propensity to operate independently from other organizations either NGO or governmental - either high or low (Curry 2010).
- •
The religious affiliation of the NGO – either affiliated with the local religious majority, minority, or secular (de Haan, 2009; Flanigan, 2010).
The case studies also informed the initial settings and ranges of the model variables in the simulation.