Implementasi Jaringan Syaraf Tiruan Backpropagation untuk Menentukan Prediksi Jumlah Permintaan Produksi Dodol Apel

Farrah Nurmalia Sari, Ari Kusumastuti, hisyam Fahmi

Abstract


Forecasting is importantly in accordance with the planning strategy; therefore it will affect the way of decision making. One of the forecasting methods is Artificial Neural Network with Backpropagation as the algorithm. This research aims to measure the accuracy of the network architecture which is being applied in order to calculate the prediction of the future’s apple paste product monthly demand which was obtained from CV. Bagus Agriseta Mandiri. The data which are being used are 36 monthly data from the year 2017, 2018 and 2019. Furthermore, the data obtained are normalized and divided into two, 66,66% as the data for training process and 33,33% as the data for testing process. Network architecture that is applied in this research is 12 : 10 :1, where 12 are neurons for input layer, 10 are neurons for one hidden layer and 1 is neuron for output layer. The Network with that framework obtained a result 20.161% for MAPE and 79.839% for the accuracy. That model is categorized as good enough for its forecasting ability. Moreover, the network was entirely validated using k-fold cross validation method with . The result obtained as follows: the average of MAPE is 47.079% and the average accuracy is 52.921%. According to it, the entire model can be categorized as good enough in order to run a forecast. As a comparison, another testing has been done with the same fold but different in the network architecture (model 6 – 8 – 1). The second model obtained results as follows: the average of MAPE is 26.74% and the average accuracy is 73.18%, so that the two prediction models’ ability are in the same category, it is good enough to run a forecast.


Keywords


Accuracy; Backpropagation; Dodol Apel Monthly Sales Data; Artificial Neural Networks; K-fold Cross Validation; Prediction.

Full Text:

PDF

References


P. Subagyo, “Forecasting Konsep dan Aplikasi Edisi 2,” Yogyakarta: BPFE, 1986.

S. P. Siregar and A. Wanto, “Analysis of artificial neural network accuracy using backpropagation algorithm in predicting process (forecasting),” IJISTECH (International Journal of Information System and Technology), vol. 1, no. 1, pp. 34–42, 2017.

R. R. P. Putri, “Implementasi Metode JST-Backpropagation Untuk Klasifikasi Rumah Layak Huni (Studi Kasus: Desa Kidal Kecamatan Tumpang Kabupaten Malang).” Universitas Brawijaya, 2018.

L. v Fausett, Fundamentals of neural networks: architectures, algorithms and applications. Pearson Education India, 2006.

Z.-H. Wang, D.-Y. Gong, X. Li, G.-T. Li, and D.-H. Zhang, “Prediction of bending force in the hot strip rolling process using artificial neural network and genetic algorithm (ANN-GA),” The International Journal of Advanced Manufacturing Technology, vol. 93, no. 9, pp. 3325–3338, 2017.

S. Chopra, D. Yadav, and A. N. Chopra, “Artificial neural networks based indian stock market price prediction: before and after demonetization,” J Swarm Intel Evol Comput, vol. 8, no. 174, p. 2, 2019.

H. Aini, H. H.-K. E. and D. Science, and undefined 2019, “Crude Palm Oil Prediction Based on Backpropagation Neural Network Approach,” journal2.um.ac.id, Accessed: Dec. 31, 2022. [Online]. Available: http://journal2.um.ac.id/index.php/keds/article/view/7335

J. J. Siang, “Jaringan syaraf tiruan dan pemrogramannya menggunakan matlab,” Penerbit Andi, Yogyakarta, 2005.

B. Widodo and S. Derwin, “Artificial Intelligence konsep dan penerapannya,” Artificial Intelligence konsep dan penerapannya, Seno, Ed. Yogyakarta: Andi, p. 7, 2014.




DOI: https://doi.org/10.18860/jrmm.v2i2.14899

Refbacks

  • There are currently no refbacks.