According to the literature, the deterioration of water quality in pipeline networks of water distribution is not solely due to the deterioration of raw water quality outlet from water treatment plants, but primarily due to the multiplication of microorganisms in water distribution pipelines, a phenomenon known as after-growth or re-growth. Presently, the most effective method of a biological stability in treated water for controlling microbial re-growth is by limiting nutrients, including nitrogen, phosphorus, and organic carbon. The content of assimilable organic carbon (AOC) within organic carbon is considered to be the most main factor for controlling the growth of microorganisms in the water distribution systems. The objects of this work were to study the Cheng Ching Lake Water Treatment Plant (CCLWTP) in Kaohsiung and the Gong Yuan Water Treatment Plant (GYWTP) in Chiayi. Water samples were collected once a month from December 2008 to November 2009. The major difference between the study objects was that front one is an advanced water treatment plant, and the other a traditional one. In order to understand the difference in biological stability between these two water treatment plants, AOC meaurement was conducted. The goals of this study were: (1) to understand the water purification process of advanced and traditional water treatment plants, and to understand the concentration in AOC fluctuation in their water distribution networks; (2) to learn about differences in how the two water treatment plants remove AOC, and to know where is improvement ; (3) to use program analysis to produce a simple formula and AOC-related water quality parameters for the two water treatment plants, providing AOC control and management strategies in the future. The results concluded that the raw water of the two water treatment plants was primarily a hybrid of hydrophobic and hydrophilic molecules, and the highest values of AOC were found in winter. The CCLWTP had an overall removal rate of 54 %, and the GYWTP had an overall removal rate of 36 %. The CCLWTP conformed to the additions of an advanced water purification unit, but the water treatment process was relatively complex. Its AOC concentration varied considerably during the course of the water treatment process, while that of the GYWTP showed more stable measurements. The CCLWTP used coagulation precipitation, rapid filtration, and biological activated carbon filtration to effectively remove the AOC. The coagulation precipitation unit used by the GYWTP was most effective process in the removal of AOC and rapid filtering was less effective one. The treated water of CCLWTP maintained an AOC concentration under 51 £gg acetate-C/L in its water distribution network, while the treated water of GYWTP mostly kept a concentration of AOC lower than 71 £gg acetate-C/L. Although the CCLWTP water pipe network had lower AOC values, it demonstrated unstable changes in levels of AOC concentration. This shows that oxidation and disinfectants in the water treatment process cannot successfully oxidize all organic matter into AOC. In contrast, the GYWTP showed a more stable removal in AOC content. For the artificial neural network system simulation, the simulation values of CCLWTP water treatment process and water distribution network are correlated less closely with the measured actual value than those of the GYWTP do. This is found to be mostly due to the relatively large fluctuations in AOC in the CCLWTP. The AOC values in the CCLWTP water treatment process and water distribution network are highly correlated to TOC, TDS, and NH3-N. For the GYWTP, AOC values were mostly correlated to TOC, temperature, and NH3-N. Finally, the two common factors for water quality at both water plants were TOC and NH3-N, we recommend that these two items can be taken into consideration to control and manage AOC in water treatment.