Abstract. Material Requirements Planning (MRP) has deficiencies when dealing with current business environments, marked by a more complex network, a huge variety of products with longer lead time, and uncertain demands. This drives Demand-Driven MRP (DDMRP) approach to deal with those challenges. DDMRP is designed to connect the availability of materials and supplies directly from the actual condition using bills of materials (BOMs). Nevertheless, only few studies have scientifically proved the performance of DDMRP over MRP for controlling production and inventory control. Therefore, this research fills this gap by evaluating and comparing the performance of DDMRP and MRP in terms of level of effective inventory in the system. The evaluation was conducted through a simulation using data from an automotive company in Indonesia. The input parameters of scenarios were given for running the simulation. Based on the simulation, for the observed critical parts, DDMRP gave better results than MRP in terms of lead time and inventory level. DDMRP compressed the lead time part from 52 to 3 days (94% reduced) and, overall, the inventory level was in an effective condition. This suggests that DDMRP is more effective for controlling the production-inventory than MRP.

This study aims to analyze the level of service quality and prioritize improvements. The method that used in this study is service quality (SERVQUAL) combined with fuzzy concepts to find out the gap between expectations and consumer perceptions through five dimensions namely tangible, reliability, responsiveness, assurance, and empathy. Fuzzy concepts are used to present uncertainties over respondents' assessment of subjective questionnaires. To determine the priority of improvement, used the Customer Value Index Potential Gains (IPGCV) method. Based on processing data results by using fuzzy-servqual method, it is known that the Tangible dimension has a gap of -0.13, the Reliability dimension has a gap of -0.13, the Responsive dimension has a gap of -0.14, the dimension of Assurance has a gap of -0.12, Empathy dimension has a gap of -0.11. The negative value on fuzzy-servqual assessment can be interpreted that the overall service has not met customer expectations. Achievement of the quality level is equal to 0.8606 which is indicates the quality of services still need improvement because the level of service quality is less than 1 (Q <1). Based on calculation using the PGCV Index, it is known that services which have the highest priority are employees who have adequate support from their institutions so that they can carry out their duties properl

Abstrak-Air minum dalam kemasan dipasarkan dengan berbagai variasi kemasan salah satunya adalah kemasan galon. Beberapa permasalahan yang terjadi pada kualitas galon adalah galon bocor, galon pecah, air kotor, body kotor dan galon berlumut. Penelitian ini bertujuan untuk menentukan performance baseline perusahaan yang dilihat dari nilai Defect Per Million Opportunities (DPMO) dan tingkat sigma, melakukan identifikasi penyebab kecacatan produk. Pembahasan yang dilakukan mengikuti langkah DMAIC pada six sigma. Berdasarkan hasil pengolahan data, dapat diketahui bahwa nilai DPMO sebesar 662,46 dan tingkat sigma sebesar 4,84. Hasil ini dapat menjadi dasar bagi perusahaan dalam upaya meningkatkan kinerja kualitas produk yang dihasilkan.Abstract--Bottled drinking water is market with a variety of packaging, one of which is gallon packaging. Some of the problems that occur in the quality of the gallon are leaky gallons, broken gallons, dirty water, dirty bodies, and mossy gallons. This study aims to determine the company's baseline performance as seen from the value of Defect Per Million Opportunities (DPMO) and sigma level, identifying the causes of product defects. Discussions were carried out following DMAIC's steps on six sigma. Based on the results of data processing, it can see that the DPMO value is 662.46, and the sigma level is 4.84. This result can be the basis for the company to improve the performance of the quality of the products produced.

Currently, data mining is being needed in various fields to obtain analysis results from certain aspects that are needed. One of the fields that make use of data mining is education. The education sector uses data mining to determine the level of performance or achievement of its students. With data mining, the education sector can also evaluate student achievement and take the next step in increasing that achievement. Data mining in education is also known as Educational Data Mining (EDM). In this study, the theme discussed was the prediction of student learning habits and steps that could be taken to improve student achievement at the university. In this study, three (3) classification algorithms were used, namely Multilayer Perceptron, Random Forest, and Support Vector Machine. This is done to find the best results from each algorithm.

The search problem is a problem that is commonly applied to systems based on the concept of Artificial Intelligence. One of the well-known heuristic search methods in Artificial Intelligence terminology is Generate and Test. In general, there are no companies operating without raw materials, raw materials in PT. DSI is a type of main and supporting raw material. Refined sugar production at PT. DSI Banten has been experiencing fluctuations in the output of production every day, the data in April 2014 showed from 1-7 consecutively that is 726, 578, 592, 518, 692, 734, 473 tons (PT. DSI, April 2014 ). The purpose of this study is to implement the heuristic search concept with the Generate and Test Algorithm in the search for a combination of the two raw materials to obtain the highest amount of production / output in the form of refined sugar, from the results of this study obtained a system that is able to find the highest amount of sugar production per cuisine, namely in the form of types of supporting raw materials (Limestone CaO, HCL, NaOH) and types of main raw materials (Raw sugar). After conducting research through the heuristic search concept with the GnT method, from 3 types of supporting raw materials (type 1: supplier from PT. SAP, type 2: supplier from PT. MNA, type 3: supplier from PT. CKT) and 3 types of raw material main (raw sugar 1: import from Australia, raw sugar type 2: import from Vietnam, raw sugar type 3: import from Thailand) found an optimization of the two raw materials with the results of supporting material type "3" and main raw material type " 2 "with the amount of 123 tons per cuisine for refined sugar output, the results obtained are able to increase productivity in the refined sugar processing.