The Gaia hypothesis, proposed by James Lovelock in 1970s, claims that the earth is a superorganism consisting of both living organisms and a non-living environment. More specifically, the earth itself has the ability to regulate climate and chemistry, making environments on earth suitable for living organisms. The Gaia hypothesis was strongly criticized by biologists, especially Darwinists. In replying to the criticism, Lovelock and Watson developed a mathematical model called Daisyworld to test the hypothesis. The Daisyworld is a planet orbiting around the sun, which is populated by two different types of plants, black daisies and white daisies. The color of the daisies influences the albedo of the planet in a way that black daisies absorb more sunlight and warm the planet, while white daisies reflect more sunlight and cool the planet. Based on this model, we analyzed the temperature-regulating ability in the Daisyworld model under various initial conditions using the methods of system dynamics. The initial conditions that proposed in this paper included three sets of initial area distribution patterns between black and white daisies and three sets of sunlight albedo combinations of the two daisies. The software package we used here is Stella 9.0.2 developed by ISEE System, Inc. The simulation results showed that, although the initial conditions of the Daisyworld model were different, all temperatures, as the system settled down, eventually reached at 21.55 ℃, a temperature close to the optimum for daisy growth. From the results we know that the Daisyworld model is insensitive to the initial conditions such as area distribution patterns and albedo combinations. While the temperature-regulating ability is insensitive to albedo combinations of the two daisies, our results showed that albedo combinations can affect the time length that the Daisyworld model needs to reach at a stable state. That is, when the albedo difference between black daisy and white daisy is smaller, the time length will be longer. Based on the simulation process and results, we further discussed the advantages and disadvantages of the Daisyworld model. We argued that simplification and systematic analysis are the advantages of the model. Through a simple feedback relationship between organisms (i.e. black and white daisies) and environment factors (i.e. temperature), the natural selection and competition between two daisies could lead to a balance of populations that tends to favor planetary habitability, without human intervention. Because of its simplification and systematic analysis, the Daisyworld model is now applied all over the world in both system engineering and environmental science. It inspires the researchers who are not used to system thinking, and changes their thinking habits forever. But this simplification has some disadvantage in the Daisyworld model. Compared with the real complicated earth system, this model is obviously too simple, which leads to a simple result. In order to get a more scientific result, we suggests that the model should include more detailed information of the earth such as more environmental variables, biological variables, feedback relationships, and human activities. The Gaia system should eventually be integrated into the more advanced Nature-Human Coupled System.