Central Sulawesi Province in Indonesia has great potential for horticultural commodities, namely local red onion Palu. In the current climate change, local farmers are still watering plants in the conventional way. The automatic watering system simplifies the work of local farmers. This device uses a soil moisture sensor as a soil moisture detector and Arduino as a program brain. This study aims to determine the position of soil moisture sensor, the optimal length of watering time and analyze the quality of data stored. The experiment was carried out using a Completely Randomized Design (CRD). The position of the soil moisture sensor was analyzed by Profile Analysis. The optimal length of watering time was determined by Analysis of Variance (ANOVA) and Least Significant Difference (LSD). The quality of data stored was determined by a number of missing values and frequency of watering. The results showed that in soil planting media the position of soil moisture sensor had no significant effect, while in others planting media (water and combination of water and soil) the position of the sensor had a significant effect. The optimal watering time was 3 seconds. The stored data has low quality in terms of missing values and lack of consistency.

Donor Darah Sukarela (DDS) adalah orang yang dengan sukarela mentranfusikan darahnya kepada orang lain. Seseorang dapat menjadi pendonor darah jika memenuhi kriteria dari PMI dan lolos dalam pemeriksaan dokter. Syarat yang diberlakukan PMI menyebabkan calon pendonor darah dapat diklasifikasikan menjadi layak dan tidak layak dalam mendonorkan darahnya. Salah satu cara untuk menentukan pola prediksi status kelayakan calon pendonor darah di PMI adalah menggunakan regresi logistik biner dan k-Nearest Neighbor (kNN). Peubah yang signifikan mempengaruhi kelayakan calon pendonor darah adalah kadar Haemoglobin. Akurasi yang dihasilkan oleh metode regresi logistik biner dan kNN pada penelitian ini adalah 93% dan 79%.

UMKM Tiga Bintang is one of the household-scale agro industries that processes roa fish, duo fish, moringa, and purple sweet potatoes as the basic ingredients into various kinds of snacks in the form of sticks. To get the best quality in accordance with the standards desired by consumers, quality control is needed, which aims to minimize deviations that are not in accordance with the standards set by the industry itself. The purpose of this study is to see whether quality control in the production process by UMKM Tiga Bintang is controlled or uncontrolled. The type of data used in this study is primary data. The analysis process carried out in this study uses the Statistical Quality Control (SQC) approach. Statistical Quality Control is a statistical technique that is widely used to control the quality of a product, by evaluating quality in terms of compliance with specified specifications. The SQC analysis results show that the number of products examined was 24800, where the average product damage was 0.095121 or 9.5121%, with the most type of damage occurring there was a brownish product colour. The upper control limit (UCL) and the lower control limit (LCL) for each production vary. Quality control of UMKM Tiga Bintang is not well controlled because of uncontrolled production time within the upper and lower control limits.

Fulfillment of rice stock in Indonesia to always be distributed based on demand in the community is certainly closely related to the results of rice production. The results of rice production in various regions of Indonesia are very different. This difference can of course be influenced by geographic location or spatial effects between regions. Central Sulawesi, which is one of the provinces with a large population compared to other provinces on the island of sulwesi, has a responsibility to meet the needs of its community, so it is necessary to take into account and increase the production of rice by relying on production in the province.Modeling of rice production that has spatial effects or heterogeneity between regions is needed as an analytical tool because if the modeling ignores spatial effects and generalizes the model, the modeling predictions will be biased. So we need an analytical model that can accommodate the problem of spatial effects using Geographically Weighted Panel Regression. The purpose of this study was to determine the factors that can affect rice production in central sulawesi. The data used comes from BPS Central Sulawesi province from 2014-2020. This study focus  to the spatial effect  factors that are considered to be able to affect the rice production production in Central Sulawesi. Tthe results of the study there area 8 districts/cities which are affected by land area, and 4 districts/cities are affected by land area and harvested are.

The physical condition of the Palu Valley shallots variety greatly affects the quality of the fried onions that are obtained. The poor quality of shallots will affect the product that will sale by the farmers. Therefore, it is necessary to monitor by conducting the quality control analysis of the physical condition of shallots. In this study we use the quality control method of the  Hotelling control map with the fast-MCD algorithm. This method is used because the outlier in the data to be analyzed. The purpose of this study is to produce average vector estimates and variance-covariance matrix estimates in the formation of the  Hotelling control map. From the calculation by using the mean vector and the variant-covariant matrix with fast-MCD estimation, 93 data were obtained that experienced out of control on the  Hotelling control map with the fast-MCD algorithm where the observations that experienced out of control were more than the usual of  Hotelling control map. This shows that the  Hotelling control map with the fast-MCD algorithm is more effective in detecting observations which contain outliers. The value of the multivariate  process capability analysis is less than one showing the process is uncapable.