طراحي مدل‌ عامل‌محور و كاربرد آن در باستان‌شناسي

Developing an Agent-Based Model for Archaeological Applications

به كارگيري كامپيوتر و دانش نرم افزاري در تجزيه و تحليل داده هاي باستان شناسي به طور چشمگيري در دهه هاي اخير افزايش يافته است. يكي از تازه هاي علوم نرم افزاري در باستان شناسي استفاده از مدلسازي عامل محور براي مطالعه ي ارتباط ميان انسان عصر باستان با محيط و آثار فيزيكي به جاي مانده از آن و آزمايش فرضيات متداول در خصوص جوامع باستاني است. سيستم هاي عامل محور معمولاً در محيط برنامه نويسي نت لوگو و يا پايتون طراحي و اجرا مي شوند. در باستان شناسي با به كارگيري دانش نرم افزاري و با استفاده از داده هاي باستان شناختي مي توان يك جامعه ي باستاني را شبيه سازي كرد و با تعريف الگوهاي رفتاري موردنظر براي عوامل تشكيل دهنده ي آن جامعه كنش و واكنش عوامل را در مواجهه با يكديگر و با محيط اطرافشان سنجيد و مورد تحليل و ارزيابي قرار داد. يكي از ويژگي هاي سيستم هاي عامل محور قابليت تعريف و اجراي چند پارامتر به طور همزمان است. به اين ترتيب مي توان براي آزمايش چند فرضيه ي متداول و مهم در خصوص يك جامعه ي باستاني چند نوع داده ي متفاوت را به صورت متغير در سيستم طراحي كرد و با تغيير دادن آن ها نتايج متفاوتي را در تحليل داده ها به دست آورد و سپس آن ها را با داده هاي واقعي مقايسه كرد. به عبارتي ساده تر با استفاده از سيستم هاي عامل محور مي توان يك جامعه ي باستاني را بازسازي كرده و با آزمايش آن با داده هاي متفاوت به روند شكل گيري يك الگوي رفتاري و اجتماعي خاص پي برد. اين مقاله قصد دارد خوانندگان و علاقه مندان را با كاربرد سيستم عامل محور در باستان شناسي و روند طراحي، ساخت و تحليل داده هاي به دست آمده از آن آشنا سازد. همچنين براي درك بهتر از چگونگي ساخت و كاركرد چنين سيستم هايي دو نمونه ي موفق از شبيه سازي عامل محور در مطالعات باستان شناسي در اين مقاله معرفي شده اند.

The aim of this paper is to consider what constitutes agent-based modelling (ABM) and how this can relate to archaeological reasoning. The development and construction of ABM models is an essential prerequisite for most archaeological reasoning. Both directly and indirectly, archaeologists are making extensive use of ideas and methods in applications that derive from archaeological, anthropological as well as ecological and behavioural theories. Technically before the development and wide spread use of computer soft wares these applications run statistical methods and are almost as varied in kind as is their quality. In applications, archaeologists today are concerned with developing or applying a model through using ABM idioms as the means for expressing relationships thought to characterize the data in question as ABM thinking is concerned with the conceptual system for which the model is constructed. The notion of an ABM model as a means to draw out what must be true in a hypothetical system through using and manipulating symbols to convey and represent concepts and properties which has been most extensively developed in the axiomatic method. Since the last decades of the twentieth century onwards there has been a rapidly growing interest in implementing agent-based modelling (ABM) in archaeological and anthropological studies. The biggest advantage of such model is creating an artificial ancient society and populating it with autonomous agents who live on spatial landscapes. In such models agents have been given certain rules of behaviour that define their way of interacting with their environment and with each other. The rules and behaviour of the agents are described by the modeller and most of the time they are stochastic. But they can also be defined based on real archaeological data and certain factors can be parameterised to test the authenticity of a hypothesis or to find out the reasons of an emergence. Emergence refers to the nonlinear relationship between multiple heterogeneous components and their interaction in their environment. Nowadays, emergence is an interesting topic to philosophy and science ranging from physics and chemistry to the social science and humanity. Our society is a good example of emergence. The creation of a society is the result of individuals’ interaction that in return influences the behaviour of those individuals. At first glance, it may seem very coherent and well integrated, but in fact it is made of a complex system in which each of us, as a component, interact with our environment through our acts and decisions. It might imply that knowing the laws of a society enable us to predict the behaviour of its components. However, even using the most scientific methodologies may not be resulted in a reliable prediction. In fact, in a complex system like a human society each component, depending on their authorities, may take decisions that result in completely different outcomes. Simulation technology is a powerful tool that enables us to predict human behaviour and decision making defined parameters. Two main potentials of ABM and simulation in archaeology are theory building and hypothesis testing. This paper is an introduction to the design and application of agent-based modelling and the analysing methods of such systems for archaeological purpose.