The aim of study is to discover outlier of customer data to found customer behaviour. The customer behaviour determined with RFM (Recency, Frequency and Monetary) models with K-Mean and DBSCAN algorithm as clustering customer data. There are six step in this study. The first step is determining the best number of clusters with the dunn index (DN) validation method for each algorithm. Based on the dunn index, the best cluster values were 2 clusters with DN value for DBSCAN 1.19 which were minpts and epsilon value 0.2 and 3 and DN for K-Means was 1.31. The next step was to cluster the dataset with the DBSCAN and K-Means algorithm based on the best cluster that was 2. DBSCAN algorithm had 37 outliers data and K-means algorithm had 63 outliers (cluster 1 are 26 outliers and cluster 2 are 37 outliers). This research shown that outlier in DBSCAN and K-Means in cluster 1 have similarities is 100%. But overal outliers similarities is 67%. Based the outliers shown that the behaviour of customers is a small frequency of spending but high recency and monetary.

Monstreation is a business engaged in clothing convection, these business products are marketed online such as jackets and shirts for class, shirt and community clothing. The problem that occurs in this convection is the lack of product recommendation services to customers. Another problem is that if there are customers who order products that are not in accordance with their needs, the customer will rarely order products at Monstreation. The solution used is to provide services that match the characteristics of the customer, for example by giving product recommendations. Product recommendations are also needed considering this type of business is a business that has many business rivals. The steps taken in this study begin by collecting customer transaction data, then the data is transformed into RFM criteria data. After being transformed, the data is weighted using AHP, after that the RFM data is weighted then grouped / clustered. The grouping results are validated with DBI. From the experiments conducted it is known that the number of cluster 3 is the optimal number of clusters in product grouping. After it is ranked based on the value of the total weight. From the experiments conducted, it is known that the results of the 3 customer clusters, the customers who have the highest weight value are customers in cluster 1. The results of this study are a product recommendation that is an association of product history of customers who have a cluster similarity and a product recommendation information system.

Abstrak— LKP Prima Tama Komputer yang saat ini sedang berjalan dalam bidang atau sektor komputer dan bekerjasama dengan Kementerian Ketenagakerjaan Republik Indonesia (Kemnaker) telah banyak menjalankan program kerja untuk pemberian beasiswa para peserta yang ingin mengembangkan profesi mereka dalam bidang komputer. LKP Prima Tama Komputer dalam kurun waktu satu tahun bisa mendapatkan dua kali program kerja beasiswa tersebut, namun permasalahan yang kini sedang dihadapi oleh LKP Prima Tama Komputer ialah sering terjadinya kesalahan dalam perhitungan data-data penilaian siswa, hal yang terjadi yaitu kesalahan pada data yang telah dirangkap oleh tim penilai namun perhitungan yang terjadi sering sekali terdapat nilai yang beda tipis hal ini menyebabkan terjadinya keraguan diantara tim penilai dengan calon siswa ditambah dengan waktu yang dibutuhkan terlalu lama adakalanya penerima beasiswa tidak sesuai dengan kenyatannya. Berdasarkan hasil dari pengujian metode SAW didapatkan nilai preferensi terbaik yaitu pada alternatif Encik Andreansyah, SY dan Cindy Maharani dengan nilai preferensi 1,96. Pada pengujian yang dilakukan menggunakan Blackbox testing, untuk semua fitur yang ada pada sistem berjalan 100% dengan keterangan sangat baik dan pengujian menggunakan metode UAT (User Acceptance Test) menunjukkan bahwa hasil penerimaan penggunan sistem ialah 92%, dan pengujian menggunakan confusion matrix didapatkan hasil 100%.Kata kunci: Sistem Pendukung Keputusan, Penerimaan Beasiswa, Simple Additive Weighting, SAW

Motorcycle users always deal with difficulties if their motorcycle tires are flat when riding in a place far from a tire repair shop. Distance is a problem that can make a flat tire quickly damaged when the motorcycle keeps forced to drive. Based on the aforementioned issues, a temporary flat motorcycle tire auxiliary tool is needed to serve as a tool that can deliver a motorcycle until it can be repaired at a tire repair shop. The method used in the design was the axiomatic design method. This method was used to determine design parameters using the mapping process. Surveys were conducted to identify customer attributes by distributing questionnaires to identify customer attributes which were then mapped into functional requirements and design parameters. The results of this study were 4 attributes based on consumer desires, among others, motorcycle wheels spinning with auxiliary wheels, easy to install, safe for bumpy roads, and robust enough to support the motorcycle body and rider.

Employee performance appraisal is needed by an agency or company with the aim of evaluating performance and improving the quality of competent human resources and high loyalty for each employee at work, then an agency or company can give awards to each of its employees such as contract extensions, salary increases , get special promotions, appointments, and allowances, which can motivate every employee. This study aims to facilitate a planner in a company PT. SUPRACO INDONESIA in providing performance appraisals of each employee uses a decision support system using the Multi Objective Optimization On The Basic Of Ratio Analysis (MOORA) method. This employee performance appraisal decision support system uses a sample of 3 employees from 11 employees using the MOORA method of calculation. the final results of the calculations carried out are: for the first rank in alternative 2 with a value of 5.7805, while the second rank in alternative 1 with a value of 5.7736, and third place in alternative 3 with a value of 5.7671. In the tests carried out using Blackbox Testing, for all the features on the system running 100% with very good information and testing using the UAT (User Acceptance Test) method, it showed that the results of system user acceptance were 92%.