Real-time Monitoring of Subsea Gas Pipelines,Offshore Platforms,and Ship Inspection Scores Using an Automatic Identification System

1 Introduction

Ship accident is fatal to the ship as well as its crew and cargo.Moreover,it has a direct impact on the sea and coastal environment.Environmental losses include the compensation for the marine industry,the costs of pollution prevention,as well as losses brought on by lost business opportunities resulting from the contamination.In 2009,the Mahkamah Pelayaran(Indonesian Maritime Court)reported 293 cases of major accidents(Mahkamah Pelayaran Indonesia 2009).Such accidents can be classified into several groups:ship sinking(31%),ship grounding(25%),ship collisions(18.27%),ship fires(9.67%),and others(16.06%).Based on reported data,the main accident causes are attributed to human(78.45%)and technical(9.67%)errors,prevailing weather conditions(1.07%),and a combination of weather conditions and technical errors(10.75%).Another study based on statistical data on maritime accidents attributed those to human error(90%)and ship collisions(30%)(Gong 2002).

Based on IMO Resolution MSC.74(69)(IMO 1998),ships with capacities exceeding 300GT are required to use an automatic identification system(AIS)in their operations.The utilization of an AIS device aims to reduce the likelihood of ship collisions occurring.An AIS system automatically sends dynamic data,such as speed,location,course,and heading,along with static data,such as length,width,draft,and other related information.At present,AIS is not only used in ships,but is also used at base stations as a vehicle monitoring system(VTS)to monitor ship traffic.VTS has been developed as intelligent,real-time multivessel collision risk assessment system to help VTS officers and ship captains during operations through the provision of real-time data(Bukhari et al.2013).

Meanwhile,the majority of Indonesia’s oil and gas infrastructure are located at sea.For this reason,offshore conventional onshore technologies are integrated in order to conduct exploration and exploitation.The technologies used comprise subsea manifolds and flow lines,wellheads,offshore platforms,export pipelines,and floating production units.

Potential hazards may occur,particularly for offshore platforms and subsea pipelines associated with ship operation while at sea and during anchorage near pipeline areas(i.e.,anchordrag,drop anchor,vessel-platform collision,ship sinking,and so on(Artana et al.2013).Past studies have proposed risk assessment methods during ship operation at sea,such as risk assessment of pipelines and subsea gas pipelines related to mooring vessel operation in tie-in spool installations(Dinariyana et al.2015),and risk assessment of subsea gas pipelines related to the development of jetties or ports(Artana 2009;Artana et al.2015).

The present research further developed an early warning system(EWS)based on a real-time monitoring system for maintaining the safety of offshore structures and subsea pipelines.In this system,the ship’s position data and dynamic AIS data are obtained in real time;hence,the relative distance between vessels and offshore platforms or subsea pipelines can also be accurately calculated in real time.The boundaries of the restricted and exclusion area of offshore platforms and subsea pipelines can be used as triggers to facilitate EWS notification.The overlay technique complements the EWS interface in such a way that vessel movement and relative distance to the offshore platforms and subsea pipelines can be visually monitored.

Visual monitoring is often required by port authorities and platform owners.One study developed an alert system called collision alert system(CAS)to prevent collision accidents in a real-time operational environment(Goerlandt et al.2015).We design a CAS asan EWS by integrating VTS/AIS with marine geographic information system to obtain the optimal decision for ship collision avoidance by using an analytical hierarchy process(AHP)and fuzzy logic theory(Su et al.2012).

In accordance with Decree No.68/2011 issued by the Ministry of Transportation of Indonesia(Menteri Perhubungan Republik Indonesia 2011),the government established several security areas for platforms,subsea cables,and subsea pipelines with specific provisions as follows:(a)restricted area for platforms,which is a region that is 500 m from the outer edge of the platform;(b)an exclusion area for pipelines and sea cables,which is located on the right and left outermost points of the pipe/cable up to a distance of 1250 m;and(c)an exclusion area for platforms,which is located 1750 m from the outer edge of the platform(or approximately 1250 m from the restricted areas.

Meanwhile,the lack of implemented safety standards is one of the major antecedents of ship accidents that occur in Indonesian seas.The Tokyo MOU Annual Report 2009(Tokyo MOU 2009)has obliged port authorities to inspect at least 25%of the total population of foreign vessels operating from local ports.

In order to conduct the inspection,port authorities require a ship inspection priority ranking system.The priority ranking inspection score is needed to prioritize the ships to be inspected,especially in ports with a high density ship population.In the current work,a ranking system using inspection scores obtained by weighing the inspection variables set by the Tokyo MOU through AHP is introduced.The inputs for the inspection score calculation include the AIS data and vessel specifications data.The calculation results are displayed via a digital map for easier visual observation.As per EWS,both the priority ranking system and ship inspection score are running in real time and are displayed on a digital map.

A real-time monitoring system is required to enhance the safety of offshore platforms and subsea pipelines as well as to reduce operational costs incurred by patrol boats when inspecting ships that have entered the restricted zone.This system is not only used for monitoring,like any other existing VTS in Indonesia,but is also used to generate early warnings for incident with high risks.The proposed system can also be integrated with all existing satellite-based VTS in Indonesia.

式中,ppd为探测器读出光功率;B为测试带宽;λ为光谐振频率;θ为探测器光电转换效率;c为光速;h为普朗克常量;D为谐振腔直径;Q为谐振腔品质因数。由式(1)可知:

2 Related Works

The distance calculation is preceded by a calculation of the angle between the locations of ship toward the ends of the pipe segment.The perpendicular formula is used to calculate the shortest distance if it is in one corner of the upright.Otherwise,the shortest distance is calculated at each end of the pipe segment(Sunday 2015).

A ll of the software listed at Table 2 are run on a 64-bit Debian 7.8 Linux(Wheezy)operating system.They come with configuration files,all of which have been fine-tuned to match the needs of execution speed,server hardware power,and computational load.Moreover,some configuration parameters are specifically set to meet the requirements of memory capacity(RAM),network configuration,variable capacity,and storage type(see Fig.3).

Table 1 AIS message specifications

Source:ITU standard(ITU-R 2014)

Type Function Remark Class A and class B shipborne mobile equipment outside base station coverage 1 Position report class A Scheduled position report;Class A shipborne mobile equipment 2 Position report class A Assigned scheduled position report;Class A shipborne mobile equipment 3 Position report class A Special position report,response to interrogation;Class A shipborne mobile equipment 18 Position report class B Standard position report for class B shipborne mobile equipment to be used instead of messages 1,2,3 19 Position report class B Extended position report for class B shipborne mobile equipment;contains additional static information 27 Long range AIS broadcast message

The above-mentioned research utilized AIS data for various assessments.Hence,the current paper presents an application of AIS data into a real-time monitoring system to track the safety of subsea pipelines as well as ship’s danger score based on academic research.

由表2中可以看出，使用BIC信息准则，得到现货与1月、2月、3月和4月期货合约下的领先、滞后阶数K分别为1、2、3和4，β的估计值分别为0.9937、0.9760、0.9624和0.9514，均非常接近1；残差ADF检验结果显示，现货分别与4个期货合约均具有协整关系。由于协整关系仅是无偏性假说成立的一个必要条件，还需要进一步检验β=1是否成立；Wald检验检验结果显示，现货与1月期合约下，β=1成立，而其他几个合约下β=1均不成立，说明原油期货市场在短期是有效的，但在长期是非有效的。

3 Prototype o f the Real-time Monitoring System for Subsea Pipelines and Offshore Platform s

The AIS consists of 27 messages,each with a different specification and utility,in accordance with the ITU standard(ITU-R 2014).

Table 1 presents AIS message types 1,2,3,18,19,and 27,which contain the ship’s latest location data.These dynamic data are translated into numerical data,such as longitude and latitude.Apart from the location data,other data are used to determine the safety levels of offshore platforms and subsea pipelines.Such data include direction(heading and course),navigation status,time stamp when transmitting data,and speed over ground.

AIS data are received and then compared with the location of the offshore platform and subsea pipeline.When the analysis results indicate that the object is in danger(i.e.,the ship is located in the limited or exclusion zone with low or zero speed,indicating that the ship tends to drop anchor),the EWS automatically sends a warning notification to all the port and ship stakeholders via SMS,big screen notification,or mobile notification(Fig.1).The system reduces the frequency of patrol boat operations,lowers total cost of pipeline safety monitoring,and provides information on the pipeline safety status to to plevel management(stakeholders).

Fig.1 Main concept of the AIS ITS software

Fig.2 Master and worker process

Each stage of the computation process,database query,requires time to complete and fall into race condition.Meanwhile,the AIS data stream is received continuously when a ship enters the reception area of the AIS device.To overcome these problems,we created several worker processes(slave program)that run in parallel with one another.Each stage of the computation process now has a set of worker processes that are controlled by one master process(controller program).Master and worker process of the AIS ITS software is shown as Fig.2.

The main task of the master process is to ensure that incoming data are processed as fast as possible and that there are enough worker processes for every computation stage.If the incoming data volume increases while all the worker processes are busy,the master process creates a new worker process to do the job.When a worker process is finished performing the assigned duties,it reports its status to the master process.This way,new incoming data can be delegated to an idle process(Fig.2).To ensure that all worker processes can perform their functions properly,we test these with dummy and real data from testing servers.By conducting the performance test,we can ensure that each worker processes’conditions simulate actual operating conditions.

Fig.3 AIS software interconnection

Table 2 Open source software specifications

Software Version Function Socat 1.7.1.3 Receives AIS data stream in the form of UDP packets from Furuno FA30 PHP-CLI 5.6.9.1 The server-side script engine;changes AIS data received from Socat into numerical data mySQL 5.6.23 Stores numerical AIS data from PHP-CLI into storage Redis 3.0.2 The data structure server;management of in-memory variables list.Redis uses PHP-CLI Mem cached 1.4.13 Key-based Memory Cache;Saves process control variable and status of computing process with PHP-CLI PHP-FPM 5.6.9.1 Server-side script engine;fetches data from mySQL database and sends it to web server Nginx Nginx 1.8.0 Web server;AIS data visualization Javascript 1.6 Client-side script engine;user interaction with the interface Mapproxy 1.8 Map tile proxy;proxifying map tile request from the client

The main idea behind the master and slave process is to overcome the race condition problem.Let us assume that there exists a job stack consisting of several jobs that must be completed by the system.If each job in the job stacks consumes one time unit,one process technique can only solve one job/time.Using the multi-process technique,we can then solve N job/time.In the PHP-CLI mode,several methods can be used to perform parallel computing,i.e.,exec,fork,and threads.In this research,exec is used because it is lightweight,simple,and does not need additional extension to install(Pitt 2015).

Fig.4 Hierarchy chart of criteria inspection score

The AIS-based EWS also utilizes some open source software available in the Internet,in addition to developing a dedicated software with PHP script.

The calculation of the inspection score begins by calculating the relative weight of each sub criteria and the criteria presented in Fig.4.Data processing by Expert Choice software is then employed to determine the priority level of each criteria and sub criteria.AHP structure models used by Expert Choice comprise an inverted tree model.Tables 3,4,5 and 6 show the AHP calculation results of Expert Choice according to answers extracted from the questionnaires accomplished by experts.

苗期田间铺设银灰色膜避蚜，每667平方米铺设银灰膜5千克或将银灰膜剪成10～15厘米宽的膜条，间距15厘米，纵横拉成网眼状。

To estimate the distribution of emissions generated by ships,Ariana et al.(2013)and Artana et al.(2014)combined the geographic identification system(GIS)and AIS data.Meanwhile,Pitana et al.(2011)utilized GIS,AIS data stream,and danger score to develop a hazard navigation map based on the weight criteria obtained from an analysis that employed the AHP.AIS has also been applied to obtain the risk of shipcollision by using fuzzy inference system(Pratiwi et al.2016;Chen et al.2014).

4 Prototype o f the Real-time Monitoring System for Ship Inspection

Inspection score is measured using the multi-criteria decision method called AHP.AHP is a technique that isemployed for measuring items using pairwise comparison and expertise of judgments to derive a priority scale.AHP is structured by multiple criteria,each with their own sub criteria(Saaty 2008).Based on the Tokyo MOU PSC(Tokyo MOU 2009),the following criteria to construct AHP affect ship inspection targets.

Table 3 Function and inspection score:flag state

Criteria Weight Function Inspection score Flag state 0.306 1000 Sub criteria:Indonesia 0.082 306 25.092 China 0.086 306 26.316 Korea 0.066 306 20.196 Panama 0.095 306 29.07 Filipina 0.151 306 46.206 Cambodia 0.252 306 77.112 Cyprus 0.062 306 18.972 Hong Kong 0.072 306 22.032 Liberia 0.086 306 26.316 Singapore 0.048 306 14.688

Table 4 Function and inspection score:classification

Criteria Weight Function Inspection score Classification 0.256 1000 Sub criteria:NK 0.038 256 9.728 LR 0.042 256 10.752 BKI 0.219 256 56.064 BV 0.057 256 14.592 GL 0.042 256 10.752 ABS 0.055 256 14.08 CCS 0.114 256 29.184 DNV 0.069 256 17.664 IRS 0.093 256 23.808 KR 0.116 256 29.696 RINA 0.087 256 22.27 RS 0.070 256 17.92

4.1 Flag State

Flag state gives a ship an identity and an issue sail perm it that complies with the law s applicable in the ship’s origin country.Flag state must ensure that a ship flying its flag is seaworthy.Among others,this factor is the most significant one.

本研究在构建生活垃圾分类投放方案评价指标体系时，对原本的分类效果评价指标进行了优化，采用基于试验分析的综合正确率这一客观指标取代公众支持率、管理执行难度、公众识别与实施难度这3个主观性较强的指标，作为分类效果的评价指标，提升了评价指标体系的客观程度。

Table 5 Function and inspection score:type of ship

Criteria Weight Function Inspection score Type of ship 0.190 1000 Sub criteria:Container ship 0.046 190 8.74 LNG 0.129 190 24.51 VLCC 0.078 190 14.82 Ferry 0.158 190 30.02 Large passenger 0.112 190 21.28 Bulk carrier 0.066 190 12.54 Leisure/fishing 0.042 190 7.98 LPG 0.139 190 26.41 PCC 0.114 190 21.66 Reefer ship 0.056 190 10.64 Towing vessels 0.060 190 11.4

Table 6 Function and inspection score:ship age

Criteria Weight Function Inspection score Ship age 0.248 1000 Sub criteria:＜5 years 0.026 248 6.448 5-10 years 0.042 248 10.416 10-15 years 0.085 248 21.08 15-20 years 0.157 248 38.936 20-25 years 0.250 248 62＞25 years 0.440 248 109.12

4.2 Classification

Ships that are 5 years or older may undergo heavier inspection.Generally,ships are designed and built for 25 years of operation.The types of inspection also depend on the ship’s age.

4.3 Ship Age

Classification is regulatory body that issues the certification of ship building and ship worthiness.Its role is significant because it lays down the ship building foundation regulated by the world’s ship safety agencies.Moreover,it also serves as a quality assurance indicator of marine insurance.

4.4 Ship Type

第二，影响研究与本质研究在时间上并行，在内容上交叉。一方面，考察明治文学创作、明治文学批评对中国作家写作与学者研究的影响。如对鲁迅、王国维的影响，比较有代表性的是王向远、潘世圣、李冬木、祁晓明等学者的论文。［36－43］值得一提并难能可贵的是，李冬木发文指出鲁迅《狂人日记》“食人意象”对日本的借鉴和模仿不久，便有李有智、王彬彬、祁晓明、周南、张志彪等学者［44－47］接连发文各抒已见参与论争，显示了我国学者认真求实的学风和对中华学术的主人翁责任感。

To assess the safety index of the ship(danger score),Pitana et al.(2011)considered other criteria apart from those mentioned above.The additional criteria for the danger score,which are obtained from AIS data,include the following:relative distance between vessels,ship speed,current,and wind,as well as the ship’s principal dimension,trajectory,and so on.

多约束条件费用模型在节水中的实践与运用…………………………………………………张浩明，邵圆媛（1.10）

Table 7 Inspection score computation example

Flag state Indonesia 25.092 Classification NK 9.728 Ship type Container ship 8.74 Ship age ＜5 years 6.448 Inspection score 50.048

Fig.5 Outer bounding rectangle(OBR)

The next process is the creation of a questionnaire for expert judgment.The portstate control officer(PCSO)is then requested to fill the form with a scale value of 1 to 9.The survey results are then processed with Expert Choice software.

Ship type identifies the function to be performed and the payload carried by the ship.Ship type affects the regulations used in buildings and types of inspections that need to be conducted.

Artana et al.(2011)utilized the same approach in determining the danger score of a ship.The same approach is utilized to accommodate the four criteria set by the Tokyo MOU to identify the need for inspection by using an index called inspection score.This is calculated using the formula:

where wi=weight of it h criteria,and fi=function values of it h criteria.

For more details,see Fig.3.

Fig.6 Ship location checking algorithm with OBR

Fig.7 Shortest distance

A single destination exists at the top of the tree;this represents the goal of the decision-making problems.The decision has a 100%weight at this point.Just below the goal are all the criteria that affect the inspection score.Based on the rating obtained,the weights of the goals should be shared between the points of criteria.Several methods in the expertchoice can be used to accomplish this task,and these are initiated by comparing all the criteria to determine the weight distribution or dissemination of such criteria.The comparison results using this software consist of the criteria with priorities taking the precedence and the value of consistency ratio.These indicate whether the value of the weighting of this risk is consistent enough to be used.

The weights of the criteria and sub criteria in Tables3,4,5 and 6 are obtained from past studies published by Artana et al.(2011)and Masroeri et al.(2012).A ll values in Tables3,4,5 and 6 are used to calculate the inspection score in real time as new AIS data captured by the AIS device.

在春季，在养殖业的过程中，鸡容易产生多种疾病。在育种过程中，只有仔细观察才能做出更好的判断和分析育种者用它做出更好的判断疾病的处理。否则，它会影响鸡的正常繁殖，并在同一时间，它会影响养鸡业的经济效益。因此，只有通过正确的判断和分析，才能采取更加针对性的预防措施来减少养殖户的损失。

For example,if a ship has the flag state of Indonesia,is a container-type ship,has an NK class,and has a ship age less than 5 years,the inspection score is computed,as shown in Table 7.

5 Prototype Development

5.1 Outer Bounding Rectangle

Fig.8 Distance of point to segment

Fig.9 Locate feature in use

The computational load of AIS data processing is influenced by several key parameters,including the number of ship,the number and complexity of the underwater object or area to be monitored,and the AIS class types.The amount of data variation is usually large,resulting in reduced server computing capacities when the calculations are done for all data(all ships and all objects).

The easiest way to limit the amount of data to be calculated is to use the basic concept of R-tree indexing.Figure 6 explains the basic concept of R-tree,which is to create an outer bounding rectangle(OBR)of each object,and then merge any overlap,adjacent,or subset of an OBR,which becomes a greater OBR(Fig.5).The OBR is not an error,it is a rectangle(also called the OBR level 2)that is created to enclose a pipeline segment to a distance of 500 mon both sides of the outer point of the pipeline(i.e.,the limited area).If two adjacent rectangles are observed,a new OBR(OBR level 1-ABCD rectangle)is formed.Hence,in using this approach,the definition of all limited/exclusion areas of the pipeline can be defined easily.Moreover,such an algorithm makes computer coding easier.

The basic concept R-tree indexing is used to determine whether a particular ship with a specific location within a specific time deserves to be a candidate data to be processed in the next computing process.This is the basic criteria used to restrict the universe of ship data.For more details,follow Fig.6.

2)数据的种类齐全。煤矿安全生产数据属于多媒体数据，不仅包括实测值、平均值、累计值等结构化数据，而且还包括矿图、图像、视频、音频、应急知识、事故案例等半结构化和非结构化数据，并且此类数据所占比例越来越大。

5.2 Calculation of the Ship Distance to Some Object

Calculating the ship distance to a certain object not only requires a simple calculation of the perpendicular distance of the ship,but must also pay attention to the location of the ship to the pipe segment.Further details are shown in Fig.7.

Fig.10 Zoom feature is being used

Fig.11 Ship info interface

For each segment,such as segment S shown in Fig.8,the shortest distance of point P to line segment is P-P(0)and not P-P(b).

In their respective works,Masroeri et al.(2012),Pitana et al.(2008,2010),and Kobayashi et al.(2010)reviewed several research that employed AIS data,including danger score,hazard navigation map,ship evacuation assessment,AIS for emission distribution and monitoring,and AIS for vessel inspection.Artana et al.(2011)combined AIS data and fuzzy clustering to measure the danger scores of ships.

5.3 Main Feature of AIS ITS

5.3.1 Location

罗瑞笑了：“那为什么管这个闲事？警察也管这些事？真是有钱能使鬼推磨呀。”他的笑容很动人，像个孩子，老福不得不承认自己有点喜欢这个小伙子。

The locate feature is employed to locate the ship in the map as well as to reposition the center of the map to the location of the ship as can be seen in Fig.9.

5.3.2 Zoom

The zoom feature is used to increase or decrease the zoom level(zoom in/out).It also repositions the center of the map to the location of the ship(see Fig.10).

国内研究大体呈现出以下局面：一、相关的学术文章数量少；二、介绍性文章居多，并主要集中在对西方翻译诗学的介绍；三、更注重翻译诗学的外围解释力，脱离诗歌翻译语境；四、对我们国家的翻译诗学理论关注不够。虽然田传茂、丁青（2006：64）对扎根于传统和实践沃土的中国当代译论话语进行了梳理，但也属于介绍性研究，所以他们在文中语重心长地指出，“原创性研究正是当前我国翻译理论研究所欠缺的。”

Fig.12 Base map

5.3.3 Ship Information

The ship information feature in Fig.11 is used to display the information window containing AIS data and ship specifications.Data shown in the info window is the latest AIS data received by the server.

Fig.13 Inspection score rank list

5.3.4 Base Map

Just like any other GIS software,the AIS ITS software also uses base map for visualizing ship location(Fig.12),offshore platform location,and subsea pipeline location.Several main sources can be used as a base map,including Open Street Map(omniscale.net),Map Quest(mapquest.com),Carto Light(cartodb.com),stamens(stamen.com),Open Street Browser(open street browser.org),Ovi Normal Day(ovi.com),and Ovi Satellite(ovi.com).

5.3.5 Inspection Score Ranking

The inspection score ranking list is a list of ships that are sorted based on the value of the inspection score.Figure 13 shows the inspection score of ships.

传统切割法需使用尖刀、止血钳、吸引器等手术器械，术中须压迫止血并结扎出血点，手术相对较繁琐，因而耗时较长。而等离子刀兼具切割和凝血功能，可在切除过程中一边切割组织，一边止血，进而有效缩短手术时间，减少出血量，且术中手术器械较少，咽腔视野相对更开阔，术中出血量相对较少，手术视野更清晰，因而有效提高手术效率，加快手术进程，减少手术时间。而传统切割法需钝性分离黏连组织，出血量较多，且无法及时止血。本研究结果显示，观察组患者的手术时间更短，术中出血量更少，与凌恺等[4]人的研究结论基本一致。

Fig.14 Vessel tracking using AIS

5.3.6 Vessel Tracking System

从表8可以看出，工作满意度的成长满意度、工作本身满意度、晋升满意度与任务绩效、人际关系、工作奉献显著相关；社会交往满意度、报酬满意度、上司满意度与工作绩效不显著性相关。整体满意度与工作绩效在0.1的显著水平下才与工作绩效显著相关。即使工作满意度与工作绩效不存在显著相关，也应该看到工作满意度会影响到员工的离职率，因此，在人力资源管理中不可以忽略员工的工作满意度，因为研究发现社会交往满意度、工作本身满意度以及报酬满意度、成长满意度几个方面也对绩效存在正向显著相关。

基于配变负载历史数据，对采样数据进行分类预测模型的选择，通过比较分类模型的优劣选出最优模型。分类器模型的选择过程以及分类器的准确率如下图所示。

In analyzing ship incidents and accidents,historical data before an incident/accident are highly needed.Assessment on risk and avoidance/mitigation effort is needed to ensure that the same accident is not repeated.The application developed in this research also provides a way to track the path of one vessel,two vessels,or a group of vessels in a certain time duration.Using this application,forensic data on ship accidents can be easily performed,and accurate analysis and recommendations can be provided.Figure 14 shows the display interface of a vessel tracking system.

6 Conclusions and Future Work

This paper presents a comprehensive technique on how to develop a real-time monitoring and EWS of offshore platforms and subsea gas pipelines using the AIS data stream.The same data and technique are then used to develop a real-time monitoring system for ship inspection by means of an index called inspection score.Both prototypes are equipped with simple features and interface,while fulfilling the basic need to increase the safety level at sea.This system can be validated to increase the accuracy of the position by directly measuring the ship distance to some object and its path in collaboration with port authorities,pipeline operators,and any related stakeholders.The system also gives broader opportunities through which to develop more advanced features in the future,which can help incoming new technology,infrastructure,and regulations.

Acknowledgments The authors are thankful to Kobe University,Japan for the AIS receiver grant provided to ITS Surabaya to conduct the research.Appreciation is also extended to the Ministry of Research,Technology,and Higher Education of Indonesia for providing the grant through the PUPT and CPPBT Program.

References

Ariana M,Pitana T,Artana BK,Dinariyana B,Masroeri A(2013)Gaussian plume and puff model to estimate ship emission dispersion by combining automatic identification system(AIS)and geographic information system(GIS).J Mar Res 3(1):1-13

Artana K(2009)Penilaian resiko pipa gas bawah laut Ujung Pangkah-Gresik dengan standard DNV RP F107.Jurnal Teknik Mesin 9(1):28-42(In Indonesian)

Artana K,Dinariyana D,Masroeri A,Pitana T(2011)Combining AIS data and fuzzy clustering to measure danger score of ships.J Mar Res 1(1):33-41

Artana K,Dinariyana A,Ariana I,Sambodho K(2013)Penilaian risiko pipa gas bawah laut.Guna Widya Publisher,Surabaya,pp 123-133(in Indonesian)

Artana K,Pratiw i E,Dinariyana DP,Pitana T,Masreori AA,Ariana M,Arifin D(2014)Enhancement on methodology for estimating emission distribution at Madura Strait by integrating automatic identification system(AIS)and geographic identification system(GIS).Proceedings of the 3rd International Symposium of Maritime Sciences,Kobe,Japan

Artana K,Handani D,Sambodho K,Pratiwi E(2015)Risiko pipa gas bawah laut akibat pembangunan dan operasional dermaga.Conference of Badan Inovasi Bisnis dan Ventura(BIBV)ITS,Surabaya,Indonesia(In Indonesian)

Bukhari A,Tusseyeva I,Lee B,Kim Y(2013)An intelligent real-time multi-vessel collision risk assessment system.Expert Syst Appl 40(4):1220-1230.https://doi.org/10.1016/j.eswa.2012.08.016

Chen S,Rashid A,Lee B,Kim D(2014)Composition ship collision risk based on fuzzy theory,vol 21.Central South University Press and Springer-Verlag,Berlin,Heidelberg,pp 4296-4302.https://doi.org/10.1007/s11771-014-2428-z

Dinariyana A,Setyorini P,Masroeri AA,Antara D(2015)Penilaian risiko pipa dan kabel bawah laut akibat penggunaan mooring vessel pada proses pemasangan tie-in spool.Proceeding of Seminar Nasional Kelautan X,Univ.Surabaya(In Indonesian)

Goerlandt F,Montewka J,Kuzmin V,Kujala P(2015)A risk-informed ship collision alert system:framework and application.Saf Sci 77:182-204.https://doi.org/10.1016/j.ssci.2015.03.015

Gong IY(2002)The development of a supporting system for safe navigation on a basis of a risk.3rd Research Report(Development of a core technology for a risk reduction)

IMO(1998)Annex 3,recommendation on performance standards for an universal shipborne automatic identification systems(AIS),IMO resolution MSC.74(69).International M aritime Organization,London

ITU-R(2014)ITU Telecommunication Standardization Sector.,available from M.1371-5,Technical characteristics for an automatic identification system using time division multiple access in the VHF maritime mobile frequency band:http://www.itu.int/dms_pubrec/itu-r/rec/m/R-REC-M.1371-5-201402-I!!PDF-E.pdf[Accessed 11 Aug 2016]

Kobayashi E,Wakabayashi N,Makino H,Ishida K,Pitana T,Hwang S(2010)Installation of an Asian AIS data receiving system network.Proceeding of Japan Institute of Navigation,Tokyo,Japan

Mahkamah Pelayaran Indonesia(2009)Rekapitulasi data kecelakaan kapal.Pusat Data dan Inform asi,Sekretariat Jenderal-Kementerian Perhubungan,Jakarta(in Indonesian)

Masroeri AA,Artana KB,Pitana T,Dinariyana D(2012)A review on some research issues on AIS to improve the ship safety operation at sea.J Mar Res 2(1):11-23

Menteri Perhubungan Republik Indonesia(2011)Alur-Pelayaran di Laut.Retrieved from dephub.go.id:http://tinyurl.com/pm-no-68-tahun-2011-pdf(in Indonesian)

Pitana T,Kobayashi E,Syunichi K,Onoda K,Rusman(2008)A large passenger ship evacuation assessment due to tsunami attack.J Jpn Soc Nav Archit Ocean Eng 8:195-207

Pitana T,Kobayashi E,Wakabayashi N(2010)Estimation of exhaust emission of marine traffic using automatic identification system data(case study:Madura Strait Area,Indonesia).OCEANS 2010 LEEE Sydney 24-27 May 2010,CFP100CF-CDR 978-1-4244-5222.Library of Congress,2009934926

Pitana T,Dinariyana A,Artana BK,Zaman M,Persada H(2011)Development of hazard navigation map by using AIS data.J Mar Res 1(1):43-52

Pitt C(2015)Multi-process PHP.Available from https://medium.com/async-php/multi-process-php-94a4e5a4be05#.py8liw fi7[Accessed 12 Aug 2016]

Pratiwi E,Artana K,Dinariyana A,Setyorini P(2016) Aplikasiautomatic identification system(AIS)untuk menentukan risk collision kapal berdasarkan fuzzy inference system.Proceeding of Seminar Nasional Kelautan X Universitas Hang Tuah,Surabaya,Indonesia(in Indonesian)

Saaty LT(2008)Decision making with the analytic hierarchy process.Int J Serv Sci 1(1):83-98.https://doi.org/10.1504/IJSSCI.2008.017590

Su CM,Chang KY,Cheng CY(2012)Fuzzy decision on optimal collision avoidance measures for ships in vessel traffic service.J Mar Sci Technol 20(1):38-48

Sunday D(2015)Lines and distance of a point to a line.Available from http://geomalgorithm s.com/a02-_lines.html[Accessed 11 Aug 2016]

Tokyo MOU(2009)Annual report on port state control in the Asia-Pacific region 2009.The Tokyo MOU,Tokyo