While the software development of a water monitoring programme is developed using the C/C++ language in Arduino Bluno. Arduino Bluno (on-board BLE chip: TI CC2540). This tool functioned as a micro-controller, which controlled the data transfer. Based on this experience, Waterdash is foresee to extend its objectives to any contaminant related to water quality. Equipped with solar and low-cost sensors behind its development, the flexibility of Waterdash features has made it is easy to use and can be personalized according to any business or research needs. With the dashboard, users do not need to manually analyze the water all the time and can monitor the status using it as a long-distance diagnostic and maintenance device.

Moreover, Waterdash module in providing predictive services can forecast contaminant of interest up to several days ahead. Various experiment and benchmarking of different machine learning and statistical time series algorithm has been experiment on the benchmark data. At last, The predictive model is developed using a deep learning algorithm with its capability to process a large volume of time series data efficiently [2]. The framework in Figure 2 is the motor behind the predictive capabilities of Waterdash. Not limited to only this type of data, users with their own business goal can upload any type of benchmark data of interest and perform analytical and predictive model [3].

Copyright reserved 2021 © Universiti Sains Malaysia