Page 203 - AI for Good-Innovate for Impact Final Report 2024
P. 203

AI for Good-Innovate for Impact



               Use case – 47: Water conservation using AI-enabled smart irrigation

               systems in agriculture Use-case                                                                      47 - UDOM














               Country: Tanzania 

               Organization: University of Dodoma 

               Contact person: Joel Emmanuel Mshana joelmsha4@ gmail .com, +255755787327

                                            Adolf Faustine adolffaustine200@ gmail .com ,+ 255 711583374/+255629653757


               47�1� Use case summary table


                Domain                            Water conservation and Agriculture
                                                  Inefficiency and water wastage associated with
                Problem to be addressed           traditional irrigation methods in Tanzania and simi-
                                                  lar regions facing water scarcity.
                                                  •  AI-driven smart irrigation systems optimize
                                                     water usage.
                                                  •  Real-time monitoring and adjustment of irriga-
                Key aspects of the solution          tion schedules.
                                                  •  Sustainable agriculture practices supported by
                                                     optimized water usage, minimizing environmen-
                                                     tal impact.

                                                  Irrigation systems, water conservation, Artificial
                Technology keywords
                                                  intelligence, IoT
                                                  Public data from ministry of agriculture in tanzania
                Data availability
                                                  and private  data
                                                  Numerical such as weather data, Soil data, EX data,
                Metadata (type of data)
                                                  plant profile data, etc.
                                                  Train - LSTM model
                Model Training and fine tuning    Fine tuning - the LSTM model can be fine tuned for
                                                  good start
                                                  Dodoma region, selected for its representative soil
                Testbeds or pilot deployments     types, crop varieties, and climate conditions preva-
                                                  lent in Tanzanian agriculture.












                                                                                                    187
   198   199   200   201   202   203   204   205   206   207   208