Production Results Forecasting Using Linier Regression Methods In UMKM KWT Suka Maju

Abstract

Kelompok Wanita Tani (KWT) SUKAMAJU is a women's association in Palihan village, Bambanglipuro sub-district. The KWT activities include monthly social gatherings, savings, and loans, gardening by growing vegetables and family medicinal plants in the KWT Sukamaju garden, regular monthly meetings for agricultural extension agents, and food production. The community service team assists in the form of increasing production by supporting production equipment which is considered to be still long for large quantities of products and multimedia and internet-based marketing as a means of establishing marketing communications for the group of craftsmen for processing banana tree food made by KWT Sukamaju. Amid the physical distancing period of the Covid-19 pandemic, community visits to the field as the only marketing communication channel must stop. The use of internet marketing means to spur production to be fast and in large numbers. The difficulty of producing such large quantities is constrained by the tools and the processing time that must be fast due to a large number of orders from the internet. Data retrieval is taken from the use of production equipment for 30 days. Every day with an intensity of 3 times every 3 until 4 hours for 6 months. The total sample is 180 data samples every three times to 540 data. The test using linear regression before and after adding equipment for the production of KWT UMKM has an average result of 56.54% of the estimate after being given additional equipment of 55.63%. The result is between estimation and reality after using production tools, after and before using the tool close to an accurate estimate of between 56, 5% with 55.6% have only a 0.9% difference in error. Based on the results of the prediction of KWT production rates using simple linear regression, it can be concluded that the addition of the data production tools used to produce has a big effect, and predictions are close to accuracy using time series data, which greatly affect the accuracy of prediction calculations.