International Journal of Marine Engineering Innovation and Research, Vol. 7(2), Jun. 2022. 59-67 (pISSN: 2541-5972, eISSN: 2548-1479) 59 Establishment of Ship Allocation Model by Using Marine Logistics Database (MLDB) Mohammad Danil Arifin 1 , Fanny Octaviani 2 (Received: 20 April 2022 / Revised: 3 June 2022 / Accepted: 4 June 2022) Abstract⎯ recently, marine big data are significantly increased. If the data are effectively analyzed, it can give an advantage, and we can harness the data that is useful for a decision-maker in maritime industries. The Marine Logistics Database (MLDB) was successfully developed in the previous studies. It was developed by integrating big data into a relational database. By utilizing the extracted data from the developed database (DB), the model of ship allocation will be established. In this study, the main purpose is to develop a ship allocation model that matches the results with the existing ship allocation. In this study, the effectiveness of the allocation model was examined by checking the port constraints, ship specification, and allocation process itself. Moreover, some simulations were executed and discussed to develop new ship allocation and analyze the effective ship specification. Keywords⎯ big data, marine logistics database, ship allocation. I. INTRODUCTION 1 In recent years, big data (BD) has gathered enormous attention from academic researchers, governments in all aspects of information, and research institutes. In the case of the maritime fields, maritime data grow exponentially with the forming of the diversity of maritime data acquisition techniques, which formed as maritime big data. Maritime big data contain a vast value and embodies an enormous academic interest that can be converted into a pack of information for everyone to learn, explore, and preserve the maritime-related field, i.e., ecology, climate, disaster, and shipping industries. For example, by observing the seismic and faulting activity data, the tsunami and undersea earthquakes can be successfully forecasted [1, 2]. The other example, by analyzing Argo data, the earth seeks to intensify the global hydrological cycle [3]. In addition, analyzing the acoustic remote sensing data of groups and species distribution can be robust scientific supporting confirmation to maintain the balance of maritime ecological [4]. In such a case, it can be realized that marine big data support warning potential problem and forecasting to help decision making. Maritime big data is used for some applications in the shipping industry, i.e., by using the BDO (Big DataOcean). It can be used to semantically enrich and link data about the maintenance schedule to identify the maintenance schedules of vessels in the data lake and visualize the impact of equipment maintenance on the vessels [5]. The system dynamic model can be established by analyzing the big data correlation such as economic growth, sea cargo movement, ship bottom, ship order, ship construction, and scrapping ship to conduct the demand-forecasting [6]. Another example is Mohammad Danil Arifin is with Marine Engineering Department, Darma Persada University, Jakarta Timur, 13450, Indonesia. E-mail: danilarifin.mohammad@gmail.com Fanny Octaviani is with Department of Naval Architecture, Darma Persada University, Jakarta, 13450, Indonesia. E-mail: fanny_octaviani@yahoo.com the visualizing or monitoring system, which is important for ship construction and the visualization system of the cutting and subassembly processes [7]. In addition, the data handling frameworks with various types of data analytics are proposed by collecting the various onboard internet of things, sensors, and acquisition systems data based on the ship performance and navigation data [8]. Using these frameworks can enhance the quality and decrease the quantity of navigation information and ship performance, and it can enhance the standard and visualize the information appropriately. In the case of the exhaust gas emission evaluation, by utilizing the big data combination of geographic information system (GIS) and automatic identification system (AIS). The ship exhaust gas emission distribution can be predicted and calculated easily [9-10]. So, the green shipping environment could be shortly achieved through these data analyses. In the previous study, to manage and utilize the availability of marine big data, such as route, port, ship, trade, and AIS data, the MLDB (Marine Logistics Database) was developed [11]. Moreover, using the extracted data from the developed database, the support system of ship basic planning also can be developed [12- 15]. In summary, marine big data is also supported in shipping industries focused on ship operation, ship construction, ship maintenance, demand forecasting, ship emission, etc. However, there is limited research that is focused on ship allocation. Therefore, in this research, we focus on establishing the ship allocation model by using MLDB. By comparing the estimation result of cargo volume with the trade volume amount from the international trade statistic database, the developed MLDB has been validated. The comparison result is shown in Figure 3. It shows that the cargo volume coverage of iron ore from Australia to Japan is 95%. At the same time, the cargo volume coverage of coal from Australia to Japan is reaching 91%. In summary, it is shown that the data extracted from the MLDB is reliable to be used.