**2. Swarm intelligence**

Amorphous computing comprises a large number of interconnected computers with low processing power, memory, and intercommunication modules. Amorphous computing is also referred to as distributed computing. Swarms are the collective name for these collections of electronic devices. When the individual agents interact locally, the computer's desired coherent global behavior results from the computer's local interactions. Although there are only a small number of misbehaving agents, and the environment is noisy and threatening, the global behavior of these enormous numbers of faulty agents is long-lasting. Therefore, randomness, repulsion, and unpredictability among agents can be used to derive Swarm Intelligence (SI), which can then be used to generate multiple solutions for a single problem. On the other hand, there are no established criteria for evaluating the performance of SIs [8].

On the other hand, SI is based on simple principles that allow it to solve complex problems with only a few simple agents. An SI feature causes coherent functional global patterns to emerge from the collective behaviors of (unsophisticated) agents interacting with their environment on a local level. SI provides a foundation for investigating collaborative (or dispersed) problem-solving approaches that do not rely on centralized control or an overarching model. SI refers to the natural or artificial behavior of decentralized, self-organized collective systems that operate on their initiative. The concept is commonly used in AI research and development. Since the early 1990s, a significant amount of effort has been expended on the solution of 'toy' and real-world problems using algorithms inspired by social insects [9].

Despite the fact that many studies on SI have been presented, there are no standard criteria for evaluating the performance of a SI system. As indexes, fault tolerance and local superiority are proposed. They used simulation to compare two SI systems in terms of these two indexes. There is a pressing need for additional analytical research.

According to the researchers, "continuum models" for swarm behavior should be based on nonlocal interactions found in biology. First, they discovered that when the density dependency of the repulsion term is greater than the density dependency of the attraction term, the swarm has a constant inner density with sharp edges, similar to what is observed in biological examples. Following that, they looked for linear stability at the swarm's borders [10].
