Hasil Pencarian  ::  Simpan CSV :: Kembali

"Turbin gas adalah suatu alat yang memanfaatkan gas sebagai fluida untuk memutar turbin dengan pembakaran internal sehingga mampu memutar generator untuk menghasilkan listrik. Turbin gas memiliki tingkat bahaya yang besar, sehingga perlu dilakukan penelitian untuk menganalisis seberapa besar potensi kegagalan komponen-komponennya. Jika sebuah mesin atau peralatan mengalami kerusakan, maka seluruh fungsi akan terhenti. Oleh karena itu, aktivitas preventive maintenace dibutuhkan untuk mencegah kerusakan dan meminimasi downtime. Tahapan penelitian ini dimulai dengan menentukan komponen kritis menggunakan diagram pareto. Kemudian memvisualisasikan data-data yang didapat. Lalu, menentukan nilai parameter shape (β), parameter scale (η), reabilitas, MTTF (Mean Time to Failure), dari komponen-komponen kritis. Terakhir merekomendasikan jadwal preventive maintenance. Dalam penelitian ini pengolahan dan analisis data dilakukan melalui Big Data Analytics menggunakan R Software diharapkan kedepannya dapat dikembangkan menjadi sebuat aplikasi yang terintegrasi untuk mengimpor dan menganalisis data historis (data base), memudahkan untuk memprediksi kegagalan secara real time, memprediksi kegagalan sebelum muncul, dan dapat mengawasi equipment secara run on live. Berdasarkan hasil pengolahan data yang telah dilakukan, ditemukan bahwa ada 8 komponen kritis, Penentuan keandalan yang dilakukan dengan bantuan R software dengan menggunakan distribusi weibull menunjukkan saat 43.830 jam operasional atau 5 tahun, komponen yang memiliki keandalan paling rendah adalah Actuator dengan nilai sebesar 0,799. Keandalan sistem pada saat 43.830 jam atau 5 tahun adalah 0,866, nilai ini digolongkan sebagai kuat. Hasil dari evaluasi nilai parameter shape (β), menunjukan 7 dari 8 komponen di kategorikan IFR (Increasing Failure Rate) kegagalan ini diakibatkan oleh beberapa faktor seperti penuaan, korosi, gesekan, sehingga di sebut fase pengausan (wearout), dan solusi yang tepat untuk membuat rekomendasi jadwal preventive maintenance dengan T=80%.

---

Gas turbine is one tool that uses gas as a fluid to turn turbines with internal combustion so that it is able to turn generators to produce electricity. Gas turbines have a high level of danger, so research needs to be done to increase the high potential level of its components. If the machine is damaged, all functions will stop. Therefore, preventive activities are needed to prevent damage and minimize downtime. The stages of this research began by determining the critical components using pareto diagrams. Then visualize the data obtained. Then, determine the value of the form parameter (β), parameter scale (η), reliability, MTTF (Mean Time to Failure), from the critical components. Last scheduled preventative maintenance schedule. In this research, processing and analyzing data done through Big Data Analytics using R Software is expected to be developed in the future into an integrated application to facilitate and analyze historical data (databases), facilitate to predict in real time, predict changes before they appear, and Can keep running equipment directly. Based on the results of data processing that has been done, found that there are 8 critical components, Determination which is done with the help of R software using Weibull distribution shows when 43,830 operational hours or 5 years, the component that adds the lowest is the Actuator with a value of 0.799. The current system value of 43,830 hours or 5 years is 0.866, this value is classified as strong. The results of the evaluation of the form parameter values (β), showed 7 out of 8 components categorized as IFR (Increased Failure Rate) this improvement was caused by several factors such as aging, corrosion, friction, so it was called the wearout phase, and the solution needed for make a preventive maintenance schedule recommendation with T = 80%."

"Dalam industri hilir migas, operasional pengolahan minyak dan gas bumi memiliki bahaya cukup tinggi terhadap kebakaran, ledakan dan hilangnya kontaminan primer atau yang sering disebut dengan Lost of Primary Containment (LOPC) pada area proses. Selain itu, operasional berkaitan dengan produktifitas dan risk management sehingga sangat penting untuk pemantauan proses secara mendalam. Aspek keandalan peralatan kritikal dalam hal ini kompresor menjadi penting untuk mencegah terjadinya unplanned maintenance yang akan berdampak pada hasil produksi minyak dan gas bumi. Kerusakan kompresor dapat disebabkan oleh suhu yang melebihi batas normal, minyak pelumas yang rendah, vibrasi dan pemisah tingkat tinggi. Dengan historikal data kegagalan yang berupa waktu kegagalan, jumlah dan indikasi kegagalan selama 5 (lima) tahun yang termasuk data terstruktur dan tidak terstruktur dari peralatan kompresor, maka identifikasi pola kerusakan dan reliability rate dapat dianalisis menggunakan metode Big Data Analytic dengan R Software di mana dilihat dari nilai regresi untuk menentukan apakah distribusi Weibull dapat diaplikasikan untuk memprediksi jadwal rutin perawatan. Dengan menggunakan analisis Weibull berbasis R Software dapat disimpulkan bahwa berdasarkan nilai keandalan, kompresor yang paling kritikal merupakan kompresor 3 karena nilai keandalan saat MTTF adalah 0.415 di mana lebih besar dari kompresor lainnya. Perawatan rutin untuk setiap kompresor dilakukan dalam jangka waktu 16 - 44 hari, perawatan rutin digunakan sebagai barrier dari kegagalan.

---

Operations of Oil and Gas Industry are related to productivity and risk management, so it is essential to monitor the process in depth. This thesis is focusing on risk management of the downstream segment on the priority element of Process Safety Management (PSM). Based on research, Mechanical Integrity is the most critical element in PSM that have to be focused. The aspect of essential reliability of equipment, in this case, the compressor becomes vital to unplanned maintenance, which will have an impact on oil and gas production. Compressor damage can be caused by temperatures that exceed normal limits, low lubricating oil, vibration and high levels of separator. Historical failure data and support that include structured and unstructured data from the reciprocating compressor approximately in 5 years will be collected. It will use to identify the damage patterns and reliability rates of the equipment. The regression value will be calculated by R as Big Data Analytics Software to determine whether Weibull distribution is sufficient. By using Weibull analysis based on R Software, it can be concluded that the compressor value, the most appropriate compressor is compressor 3 because the value agreed upon when MTTF is 0.415 which is greater than other compressors. Routine maintenance for each compressor is carried out within 16 - 44 days, routine maintenance is used as a barrier from failure. "

"[ABSTRAKPT. XYZ merupakan perusahaan telekomunikasi di Indonesia yang sedangberusaha mentransformasikan bisnisnya menuju layanan broadband dan bisnisdigital. Banyak peluang bisnis di layanan broadband dan bisnis digital yang dapatdiidentifikasi dengan memproses dan menganalisis data dengan cepat, tepat, danmenyeluruh. Saat ini PT. XYZ telah memiliki kemampuan dalam mengolahbeberapa sumber data yang terstruktur dengan ukuran data yang terbatas. Untukmembuat perhitungan dan keputusan yang jitu, terutama di layanan broadband danbisnis digital, PT. XYZ dituntut juga untuk bisa memproses dan menganalisis datayang memiliki karakteristik 3V (Velocity, Volume, Variety) atau dikenal dengan bigdata. Penelitian ini bertujuan untuk merancang arsitektur sistem pemrosesan bigdata di PT. XYZ. Kerangka arsitektur (framework) enteprise yang digunakan dalampenelitian ini adalah TOGAF. Hasil yang diperoleh dari penelitian ini adalahrancangan arsitektur sistem pemrosesan big data yang mampu mengolah data yangmemiliki karakteristik 3V, yaitu aliran data yang cepat, berukuran masiv, danberanekaragam (terstruktur maupun tidak terstruktur) dengan biaya lebih rendahdari sistem pemrosesan data yang dimiliki PT. XYZ saat ini. Saran untuk penelitianini kedepannya adalah sistem pemrosesan big data di PT. XYZ dapatdiimplementasikan dengan baik jika mendapat dukungan penuh dari manajemenperusahaan, dimulai dengan kasus bisnis yang spesifik (specific business case) yangingin disasar. Hasil yang maksimal dari kasus bisnis tersebut dapat dijadikanlandasan untuk investasi sistem pemrosesan big data yang lebih menyeluruh dalammendukung transformasi bisnis menuju layanan broadband dan bisnis digital.

---

ABSTRACTPT. XYZ is a telecommunication company in Indonesia which is transforming it's business to broadband services & digital business. Many business opportunities in broadband services & digital business can be identified by processing and analyzing data quickly, accurately, and completely. Right now PT. XYZ has the capability in processing some structured data sources with limited data size. To make accurate calculations and decisions, especially in broadband services and digital business, PT. XYZ also required to be able to process and analyze the data that has the characteristics of 3V (Velocity, Volume, Variety) or known as big data. This research aims to design the architecture of big data processing system. The enterprise architecture framework used in this study is TOGAF. The results obtained from this study is the design of big data processing system architecture that is capable of processing data which has the characteristics of 3V (the fast dataflow, massive data size, and diverse structured or unstructured data sources) at a lower cost than the current data processing system in PT. XYZ. The suggestion about this study is the big data processing system can be implemented properly in PT. XYZ with the full support of the PT. XYZ management, started with a specific business use case that want targeted. The maximum results from the business use case can be used as a piloting for big data processing system investments morethorough in supporting business transformation toward broadband services and digital business. ;PT. XYZ is a telecommunication company in Indonesia which is transforming it?s business to broadband services & digital business. Many business opportunities inbroadband services & digital business can be identified by processing and analyzing data quickly, accurately, and completely. Right now PT. XYZ has the capability inprocessing some structured data sources with limited data size. To make accuratecalculations and decisions, especially in broadband services and digital business,PT. XYZ also required to be able to process and analyze the data that has thecharacteristics of 3V (Velocity, Volume, Variety) or known as big data. Thisresearch aims to design the architecture of big data processing system. Theenterprise architecture framework used in this study is TOGAF. The resultsobtained from this study is the design of big data processing system architecturethat is capable of processing data which has the characteristics of 3V (the fast dataflow, massive data size, and diverse structured or unstructured data sources) at alower cost than the current data processing system in PT. XYZ. The suggestionabout this study is the big data processing system can be implemented properly inPT. XYZ with the full support of the PT. XYZ management, started with a specificbusiness use case that want targeted. The maximum results from the business usecase can be used as a piloting for big data processing system investments morethorough in supporting business transformation toward broadband services anddigital business. , PT. XYZ is a telecommunication company in Indonesia which is transforming it’s business to broadband services & digital business. Many business opportunities inbroadband services & digital business can be identified by processing and analyzing data quickly, accurately, and completely. Right now PT. XYZ has the capability inprocessing some structured data sources with limited data size. To make accuratecalculations and decisions, especially in broadband services and digital business,PT. XYZ also required to be able to process and analyze the data that has thecharacteristics of 3V (Velocity, Volume, Variety) or known as big data. Thisresearch aims to design the architecture of big data processing system. Theenterprise architecture framework used in this study is TOGAF. The resultsobtained from this study is the design of big data processing system architecturethat is capable of processing data which has the characteristics of 3V (the fast dataflow, massive data size, and diverse structured or unstructured data sources) at alower cost than the current data processing system in PT. XYZ. The suggestionabout this study is the big data processing system can be implemented properly inPT. XYZ with the full support of the PT. XYZ management, started with a specificbusiness use case that want targeted. The maximum results from the business usecase can be used as a piloting for big data processing system investments morethorough in supporting business transformation toward broadband services anddigital business. ]"

"ABSTRAKPenelitian ini ditujukan untuk menentukan kepemilikan produk yang mempengaruhi peningkatan Product Holding Ratio PHR untuk identifikasi nasabah potensial, menentukan segmen yang terbentuk dari nasabah potensial, dan penawaran produk yang sesuai untuk setiap segmen yang terbentuk. Dari data nasabah Bank XYZ, dilakukan regresi logistic dan scoring untuk mendapatkan model dan insight untuk meningkatkan kepemilikan produk PHR serta analisis klaster untuk menentukan segmen nasabah potensial. Hasil analisis menunjukkan pengelompokan karakteristik nasabah potensial menjadi 3 tiga , yakni nasabah student, housewife, dan professional. Masing-masing kelompok memiliki profil, produk yang telah dimiliki dan kecenderungan bertransaksi di merchant tertentu. Hal ini akan dimanfaatkan untuk menyusun produk yang sesuai untuk ditawarkan dan cara komunikasi pemasaran yang tepat sesuai dengan karakteristik tiap kelompok.

---

ABSTRACTThis study is aimed to determine the ownership of products that affect the increase in Product Holding Ratio PHR for the identification of potential customers, determining the segments formed from potential customers, and offering appropriate products for each segment. Base on customer rsquo s data from Bank XYZ, logistic and scoring regressions are applied to obtain models and insights to improve product ownership PHR as well as cluster analysis to determine the segment of potential customers. The results of the analysis show the grouping of potential customer characteristics into 3 three , student, housewife, and professional customers. Each group has a profile, produc that has been owned and the tendency to make transactions in certain merchants. This will be utilized to develop appropriate products to offer and the right way of marketing communications based on the characteristics of each group."

"Pemanfaatan Palm Oil Biodiesel mulai berkembang sejalan dengan pelaksanaan kebijakan mandatori BBN dari 2.805 Ribu kL di 2013 menjadi 4.706 Ribu kL pada 2018. Keberhasilan transisi energi melibatkan sektor keuangan dan keseimbangan tiga dimensi energi yaitu keamanan, kesetaraan, dan keberlanjutan lingkungan. Diperlukan teknologi Big Data untuk menangani berbagai variasi dan komparasi data yang disesuaikan dengan parameter Keuangan dan Trilemma Energi pada penerapan Palm Oil Biodiesel. Tahapan penelitian ini melalui proses pengelompokan data numerik dan kategorik yang kemudian melewati tahap training dan pengujian data. Output nya berupa hasil studi Big Data pada risiko kebijakan energi dalam penerapan Palm Oil Biodiesel. Simulasi tersebut dilakukan dengan menggunakan pemprograman Python. Badan Usaha BBN (SMAR.JK, TBLA.JK, dan CEKA.JK) menghasilkan Stock Return sebesar 28%. Kovarians terkecil terjadi antara saham SMAR.JK dan TBLA.JK, dengan korelasi dari ketiga saham tergolong rendah. Sharpe Ratio ketiganya menghasilkan 0,83-0,84. Skenario penerapan Biodiesel ini memberikan dampak progresif terhadap pertumbuhan energy equity dan energy security namun terjadi arah penurunan pada aspek environmental sustainability dengan penjabaran tahun 2020 aspek energy equity; environmental sustainability; energy security [0,72; 0;65; 0,79] diproyeksikan tahun 2023 [0,83; 0,53; 0,83]. Dalam penerapan kebijakan energi dalam penerapan Palm Oil Biodiesel tersebut terdapat risiko yang perlu diperhatinkan terhadap pelaksanaannya. Pada krisis tahun 2008 yang menunjukkan ketidaksbilan pada sektor keuangan. Dari Grafik 4.1. Total Badan Usaha BBN Stock Return menunjukkan SMAR.JK mengalami krisis paling berat. TBLA.JK paling stabil dan CEKA.JK tidak mengalami penurunan yang signifikan. Ketika DMO (Domestic Market Obligation) Palm Oil tidak terpenuhi maka biaya lingkungan tidak tercapai

---

The utilization of palm oil biodiesel begins to develop rapidly in conjunction with the implementation of the biofuel mandatory policy, from 2,805 thousand kL in 2013 to 4,706 thousand kL in 2018. Achieving the transition involves the financial sector and the balance of three energy dimensions: security, equity, and environmental sustainability. Big Data Technology is required to handle numerous data variations and comparisons adjusted with the Finance and Energy Trilemma parameter in palm oil biodiesel implementation. This research begins with grouping numerical and categorical data and then continues to data training and data testing. The output will be the results of Big Data analysis on the risks of energy policy in palm oil biofuel implementation. Python software is used to perform the simulation. Biofuel companies (SMARJK, TBLA.JK, and CEKA.JK) yield 28% stock return. The smallest covariance exists between the stocks of SMAR.JK and TBLA.JK, while the correlation between the three companies’ stocks is considered low. The Sharpe Ratio of the three ranges from 0.83 to 0.84. This biodiesel implementation scenario contributes a progressive impact on the growth of energy equity and energy security but lowers environmental sustainability. The value of energy equity, environmental sustainability, and energy security in 2020 is expressed consecutively as [0.72, 0.65, 0.79], which is projected to 2023 with the value expressed as [0.83, 0.53, 0.83]. Risks in conducting the energy policies of palm oil biodiesel should be considered. The crisis that occurred in 2008 displayed instability in the financial sector. Based on Chart 4.1. Total Stock Return of Biodiesel Companies, SMAR.JK suffered from the crisis the most. TBLA.JK was the most secure, and CEKA.JK did not experience a significant decline. If the DMO (Domestic Market Obligation) of palm oil is not met, the environmental cost will not be attained."

"Dengan ketatnya persaingan pada Industri asuransi kendaraan bermotor dan besarnya kemungkinan hilangnya pelanggan (customer churn), perusahaan dituntut untuk semakin memperhatikan pengelolaan hubungan pelanggan. Faktanya mempertahankan pelanggan dinilai lebih efisien secara sumber daya dibanding jika harus mencari pelanggan baru. Dengan perkembangan teknik pengelolaan (CRM) dan besarnya pertumbuhan data yang ada, kemungkinan untuk menganalisa perilaku pelanggan dan mengurangi terjadinya churn dapat dilakukan dengan analisa big data dan machine learning. Dalam penelitian ini, algoritma machine learning dalam membuat model terbaik yang dapat memprediksi churn pada pelanggan. Data yang digunakan merupakan 27.013 data transaksi pembelian dari PT. Asuransi XYZ yang merupakan salah satu perusahaan asuransi otomotif terbesar di Indonesia. Setelah melakukan eksplorasi dan descriptive analytics, peneliti membandingkan akurasi antara logistic regression, decision tree, dan random forest dalam memprediksi churn pada pelanggan. Ditemukan bahwa random forest memiliki skor akurasi total terbaik yaitu 87.74% diikuti dengan decision tree dan logistic regression. Lebih lanjut lagi, tenure, presentase komisi, dan jumlah premi adalah feature yang paling penting dari model. Hasil dari penelitian diharapkan dapat membantu perusahaan dalam mengurangi churn pada pelangan dengan memprediksi lebih awal dan membuat kampanye retensi pelanggan yang tepat

---

With the tightness of the competition in auto insurance market and the possibilities of company loss because of customer churning, insurer company should put more concern in customer relationship management. Moreover, there is fact of acquiring new costumers is more expensive than retaining one. Fortunately, due to current advancement in CRM and vast growth of data provided, the possibilities to learn customer behavior from the data is becoming feasible to retain the customer. This research purpose is to used machine learning algorithm to build model provide best accuracy to predict customer churn. We use 27.013 transaction data from XYZ Insurance Ltd, one of major auto-insurer in Indonesia. After exploring the data with the descriptive analysis, we conduct predictive analysis model with logistic regression, decision tree, and random forest to determine which algorithm give the best accuracy. This research found that random forest model has highest accuracy score (87,74%) followed by decision tree and logistic regression. Furthermore, tenure, percentage of commision, and amount of premium are the most feature that impacting customer churn. This result could help insurer company to reducing customer churn by creating right marketing campaign to retain customers."

"The book describes the emergence of big data technologies and the role of Spark in the entire big data stack. It compares Spark and Hadoop and identifies the shortcomings of Hadoop that have been overcome by Spark. The book mainly focuses on the in-depth architecture of Spark and our understanding of Spark RDDs and how RDD complements big data’s immutable nature, and solves it with lazy evaluation, cacheable and type inference. It also addresses advanced topics in Spark, starting with the basics of Scala and the core Spark framework, and exploring Spark data frames, machine learning using Mllib, graph analytics using Graph X and real-time processing with Apache Kafka, AWS Kenisis, and Azure Event Hub. It then goes on to investigate Spark using PySpark and R. Focusing on the current big data stack, the book examines the interaction with current big data tools, with Spark being the core processing layer for all types of data.The book is intended for data engineers and scientists working on massive datasets and big data technologies in the cloud. In addition to industry professionals, it is helpful for aspiring data processing professionals and students working in big data processing and cloud computing environments."

"Energi listrik merupakan kebutuhan yang sangat vital bagi kehidupan manusia. Tanpa adanya listrik berbagai aktivitas tidak dapat dilakukan. Kebutuhan akan energi listrik akan terus meningkat dari waktu ke waktu. Peningkatan akan kebutuhan energi listrik ini seiring dengan pembangunan yang terjadi. Tak terkecuali di Universitas Indonesia. Pembangunan besar-besaran yang terjadi sejak tahun 2010 hingga tahun 2025 membuat kebutuhan akan listrik di Universitas Indonesia meningkat drastis. Namun, jumlah daya yang ada di seluruh gardu listrik di Universitas Indonesia masih jauh dari cukup untuk memenuhi kebutuhan tersebut. Untuk itu, dalam skripsi ini, penulis mencoba memanfaatkan potensi yang ada di Universitas Indonesia untuk memenuhi kebutuhan listrik dalam kampus secara mandiri dan melakukan studi perancangan pembangkit listrik tenaga gas sebagai alternatif untuk memenuhi kebutuhan listrik untuk Universitas Indonesia. Alternatif untuk memenuhi kebutuhan listrik tambahan di Universitas Indonesia adalah dengan membangun PLTG yang menggunakan sistem pendinginan udara masuk kompresor Absorption chiller untuk meningkatkan efisiensi dari Pembangkit. Di dalam tulisan ini juga dipaparkan analisis finansial apabila menggunakan PLTG mandiri.

---

Electrical energy is a vital necessity for human life. Without electricity, activities can not be done. The necessity for electrical energy will increase continously over time. Increased of electrical energy necessity is in line with the development, include at University of Indonesia. Massive development that have occurred since the year 2010 to 2025 made the necessity of electricity has increased significantly. However, the amount of power that exist around the electrical substation at the University of Indonesia is still far from enough to supply those necessity.

Therefore, in this last project, the authors tried to use all of pontencies to fulfill electricity necessity in this Campus and make design study of gas power plants as an alternative to supply the electricity necessity for University of Indonesia. The alternative to fulfill electricity necessity in University of Indonesia is built Gas Power Plant that used Inlet Air Cooling System to Compressor to increase the efficiency of Power Plant. In this paper is explained financial analyze too if using Independent Gas Power Plant."

"Konsumsi energi listrik di kampus Universitas Indonesia mengalami peningkatan setiap tahunnya, ini terjadi karena peningkatan pembangunan gedung. Salah satu upaya untuk memenuhi kebutuhan ini adalah dengan adanya penambahan salah satu sistem pembangit yang disesuaikan dengan kondisi lingkungan dan potensi yang ada di Universitas Indonesia. Universitas Indonesia telah tersedia pipa gas dan potensi air danau untuk pendinginan pada sistem pembangkit. Sehingga pembangkit listrik tenaga gas tepat sebagai solusi. Untuk menaikan daya output agar sesuai kemampuan original dari manufaktur turbin gas, yakni dengan cara menurunkan suhu udara masuk ke kompresor/turbin gas tersebut, sehingga perancangan PLTG dengan mechanical refrigeration dirasa tepat dalam melengkapi solusi tersebut. Tulisan ini akan memaparkan rancangan dari pembangkit listrik tenaga gas yang dapat membangkitkan daya hingga 24 MW menggunakan software Cycle ? Tempo 5.0 Pada tulisan ini pula didapat analisis mechanical refrigeration, heat balance, kebutuhan bahan bakar pembangkit, nilai efisiensi, nilai heat rate, dan analisis finansial dari pembangunan pembangkit listrik mandiri untuk Universitas Indonesia.

---

Electrical energy consumption at the University of Indonesia has increased every year, this occurs because of an increase in the construction of the building. One effort to fulfill this need is with the addition of power plant systems adapted to the environmental conditions and the potential in University of Indonesia. University of Indonesia's environment has provided gas pipeline and potential lake water for cooling the power plant system. So that proper gas power plants as a solution. To increase the power output to match the original capability of manufacturing gas turbines, namely by lowering the temperature of the inlet air to the compressor / gas turbine, so that the design of the power plant Mechanical refrigeration is appropriate in completing the solution.

This paper will describe the design of a gas power plant that can generate power up to 24 MW using software Cycle - Tempo 5.0 This paper also analyzes obtained Mechanical Refrigeration, heat balance, the needs of fuel, the efficiency, heat rate value, and financial analysis independent of the power plant to the University of Indonesia."

"ABSTRAKPT. Citilink Indonesia Citilink adalah maskapai penerbangan yang menerapkan konsep penerbangan berbiaya murah atau Low Cost Carrier LCC . Dalam menjalankan proses operasionalnya, Citilink memiliki masalah terkait efisiensi bahan bakar. Salah satu penyebab bahan bakar tidak efisien adalah tidak teridentifikasinya komponen mesin pesawat yang kinerjanya sudah tidak maksimal. Komponen pesawat yang tidak bekerja maksimal tidak dapat teridentifikasi dikarenakan data log pesawat yang terlalu banyak belum dapat diolah. Berangkat dari masalah tersebut, Citilink dituntut untuk dapat mengolah dan menganalisis data log pesawat tersebut beserta dengan data lainnya yang memiliki karakteristik Big Data Volume, Velocity, dan Variety . Namun agar penerapan Big Data berjalan dengan baik dan tepat sesuai dengan karakteristik perusahaan, maka diperlukan sebuah kerangka kerja penerapan Big Data serta memvalidasi kerangka kerja tersebut di Citilink. Metode yang dipakai untuk merancang rekomendasi kerangka kerja penerapan Big Data adalah metode penelitian Lakoju 2017 , Tianmei dan Baowen 2007 , dan Kitsios dan Kamariotou 2016 yang terdiri dari tahap rencana pemetaan, penilaian framework Big Data, perancangan framework Big Data, dan uji validasi. Hasil penelitian menunjukkan lima fase kerangka kerja penerapan Big Data Big Data Strategic Alignment, Team, Project Plan, Data Analytics, dan Implementation yang dikelilingi oleh tahap proses manajemen kinerja serta adanya keterlibatan proses manajemen perubahan. Setelah itu, kerangka kerja tersebut divalidasi kepada dua responden dari Citilink dan dilakukan enam perbaikan, yaitu penambahan cara menganalisis kondisi internal organisasi, pengkategorian posisi tim, penambahan rekomendasi data processing architecture, penambahan aktivitas pemeliharaan aplikasi Big Data, penambahan bagaimana cara menganalisis perubahan, dan penambahan aktivitas kontrol proyek.

---

ABSTRACTPT. Citilink Indonesia Citilink is airline that implements Low Cost Carrier LCC . To perform its operational processes, Citilink has problems related to fuel efficiency. One of the causes inefficient fuel is unidentified damaged aircraft engine components. An unidentified damaged aircraft components because the data log plane is too much and can not be processed. From these problems Citilink also required to be able to process and analyze the log data of the aircraft and other data that has characteristics of Big Data Volume, Velocity, and Variety . But in order for Big Data implementation to run properly and appropriatley in accordance with the characteristics of the company, it would require a framework for implementation of Big Data and validate that framework in Citilink. The method used to design the recommendation of framework for implementation of Big Data is Lakoju 2017 , Tianmei and Baowen 2007 , and Kitsios and Kamariotou 2016 research method begins with the mapping plan stage, the Big Data framework assessment, the Big Data framework design, and the validation test. The results is the five phases in framework for implementation of Big Data Big Data Strategy Alignment, Team, Project Plan, Data Analytics and Implementation surrounded by performance management processes and the involvement of the change management process. After that, the framework has been validated to two respondents from Citilink and has been improvements with adding of how to analyze an organization 39;s internal conditions, categorizing position of the team, addition of a data processing architecture recommendations, additional Big Data maintenance aspects, addition of how to analyze changes, and addition of project control aspects."