Decision Support Systems in Fisheries and Aquaculture: A systematic review

by   Bjørn Magnus Mathisen, et al.

Decision support systems help decision makers make better decisions in the face of complex decision problems (e.g. investment or policy decisions). Fisheries and Aquaculture is a domain where decision makers face such decisions since they involve factors from many different scientific fields. No systematic overview of literature describing decision support systems and their application in fisheries and aquaculture has been conducted. This paper summarizes scientific literature that describes decision support systems applied to the domain of Fisheries and Aquaculture. We use an established systematic mapping survey method to conduct our literature mapping. Our research questions are: What decision support systems for fisheries and aquaculture exists? What are the most investigated fishery and aquaculture decision support systems topics and how have these changed over time? Do any current DSS for fisheries provide real- time analytics? Do DSSes in Fisheries and Aquaculture build their models using machine learning done on captured and grounded data? The paper then detail how we employ the systematic mapping method in answering these questions. This results in 27 papers being identified as relevant and gives an exposition on the primary methods concluded in the study for designing a decision support system. We provide an analysis of the research done in the studies collected. We discovered that most literature does not consider multiple aspects for multiple stakeholders in their work. In addition we observed that little or no work has been done with real-time analysis in these decision support systems.


page 1

page 2

page 3

page 4


A Systematic Mapping on the use of Visual Data Mining to Support the Conduct of Systematic Literature Reviews

A systematic literature review (SLR) is a methodology used to find and a...

Choreography in the embedded systems domain: A systematic literature review

Software companies that develop their products on a basis of service-ori...

A Survey of Operations Research and Analytics Literature Related to Anti-Human Trafficking

Human trafficking is a compound social, economic, and human rights issue...

From Monolithic Systems to Microservices: An Assessment Framework

Context. Re-architecting monolithic systems with Microservices-based arc...

Tolerance in Model-Driven Engineering: A Systematic Literature Review with Model-Driven Tool Support

Managing models in a consistent manner is an important task in the field...

What is Legitimate Decision Support?

Decision support is the science and associated practice that consist in ...

Intelligent Transportation Systems With The Use of External Infrastructure: A Literature Survey

Increasing problems in the transportation segment are accidents, bad tra...

1 Introduction

Decision makers within the Fishery and Aquaculture111For this study we define “aquaculture” as the area dealing with cultivation and farming of aquatic biomass, whereas “fisheries” refer to traditional fishing e.g. trawl and line vessels. Industry are facing complex decision problems every day. Where to trawl with the boat tomorrow? Where do I apply to build my new aquaculture installation? These are just two examples, but answering these questions relies on knowledge from many different fields of science and heterogeneous data from many different sources. Answering where should the boat trawl tomorrow? relies on the fields of meteorology, fish biology, economics, ocean modeling and more. In addition it could require data from previous trawls (location, time and amount), weather forecast, ocean model output, fish stock models and previous logs of fuel consumption. To come to a decision in the face of all these variables and knowledge is very demanding, but doing it successfully can help decision makers optimize the operation of the business.

Systems designed for enabling decision makers to make informed decisions in the face of a complex problem are called Decision Support Systems (DSS). DSSs tries to combine domain and technical knowledge and package it in a way that can be of practical use for non-scientistsLannan (1993).

Decision support systems (DSS) stems from the 1960s Power (2008), and has been applied to a multitude of domains. Although the taxonomy and general process of creating, using and maintaining DSSs are well documented both in case studies and research, the literature provides little information regarding empirical assessments of its effectiveness in particular domains. We could only find a few systematic literature review of decision support systems research. However these were exclusively within the domain clinical decision support systems within medicine has been subject to systematic reviews targeting effectsGarg et al. (2005); Hunt et al. (1998) and how to improve such systemsKawamoto et al. (2005). More closely related to research on DSS in fishery and aquaculture is research on spatial DSS, which has been studied in non-systematic surveysCrossland et al. (1995); Malczewski (2006)

The primary research hypothesis of this paper postulates there is little empirical knowledge of the effectiveness of decision support based system in the fishery and aquaculture domain. In addition, it is believed that there is no single system that combines both their respective requirements to maximize economic efficiency, sustainability, produce optimal vessel-scheduling and reduce environmental footprints, based on multiple data sources. Therefore this study aims to aggregate primary empirical studies in an objective manner to refute or support the primary hypothesis.

Context: There are little to no decision support system for fisheries and aquaculture that supports a variety of fish species. Neither does there exist a fully integrated information system with near real-time advanced analysis models for fisheries.

Objectives: To conduct a mapping study to survey existing research on DSSs in order to identify useful approaches and clarify needs for further research. Method: A systematic mapping study of the available literature following the best practice methods laid out by previous practitionersKitchenham and Charters (2007).

Results: 27 papers have been identified by topic, system classification, and relevance for the fishery and aquaculture domain. We found that fishery- and aquaculture decision support systems rarely evaluate their system empirically, which indicates that additional investigation, empirical and practical, should be performed. In addition, the study found no DSS contradictory to the context of this study; however it identifies key methodology and insights for future usage in fishery- and aquaculture DSSs.

Conclusions: The majority of fishery- and aquaculture decision support systems published over the last 25 years focus on singular topics, and these systems does not provide multiple aspects for multiple stakeholders on the consideration of multiple factors. We observed empirical evaluation and real-time data analytics to be virtually non-existent in the problem domain.

2 Method

To gather data on the state of decision support system research within the domain of Fisheries and Aquaculture we apply a systematic literature review. This way we comply to a well known and defined method, providing reproducability and rigor while at the same time acquiring knowledge about the field and answering our research questions.

This mapping study has been conducted in compliance with a pre-defined protocol created for this study to reduce the possibility of researcher bias Kitchenham and Charters (2007) . The review protocol is an essential component for providing context and domain classification, and a protocol must be developed separately for each mapping study in order to define the main guidelines for conducting systematic mapping studies. Both Kitchenham Kitchenham and Charters (2007) and Budgen et al. Budgen et al. (2008) states that the research questions in mapping studies are likely to be broader than in traditional Systematic Literature Reviews (SLR), to adequately address the wider scope of the study. Kitchenham Kitchenham and Charters (2007) also states that mapping studies will likely return a very large number of studies which in turn will give a much broader coverage than the outcome of the SLR. On the basis of the preceding statements, a systematic mapping study was selected as the method for achieving a broad resolution on the research questions as opposed to an SLR.

The following four research questions (RQs) were formulated in order to characterize the field of DSS within fishery and aquaculture:

  • What decision support systems for fisheries and aquaculture exists? (RQ1)

  • What are the most investigated fishery and aquaculture DSS topics and how have these changed over time? (RQ2)

  • Do any current DSS for fisheries and aquaculture use real-time analytics? (RQ3)

  • Do DSSes in Fisheries and Aquaculture build their models using machine learning done on captured and grounded data? (RQ4)

2.1 Rigor of Study

The study has been conducted by three of the authors in collaboration, with two of the authors acting as a reviewers of the process. The search process of the mapping study has been executed analogous to traditional SLR studies as similar processes for searching are explicitly defined in the research protocol and reported as part of the outcome Budgen et al. (2008). In compliance with Kitchenham Kitchenham and Charters (2007), researchers should specify their rationale for the use of electronic or manual searches or a combination of both. Although most text books emphasize the use of electronic search procedures, they are not usually sufficient by themselves, and some researchers strongly advocate the use of manual searches (e.g. Kitchenham and Charters (2007)). The results presented in this study is a combination of both techniques, and is justified because the field of maritime DSS with real-time analytics has not been researched sufficiently to aggregate enough results through only automatic searches. The following sources were used for this study:

  • IEEE Explore

  • CM Digital Library

  • Google Scholar

  • Citeseer library

  • Springer

  • Ei Compendex

These sources were selected because they are among the most important repositories for acquiring data in computer science and collectively they addressed the main digital libraries deemed appropriate for this study. No researchers were contacted directly in this survey. The results retrieved from executing a search on the various data sources were either dismissed or accepted into the primary study selection process, based on inclusion- and exclusion criteria. The inclusion- and exclusion criteria are used to exclude papers that are not relevant to answer the research questions, and are one of the activities in a mapping study near identical to SLR Petticrew and Roberts (2008). The study used the following inclusion criteria:

  1. Only studies written in English or Norwegian

  2. Studies noting or referencing any of the subjects described in the research question in their title or abstract

  3. Studies had to be published after 1990 222And not after the first half of 2015 as this is when this study finished.

  4. Studies had no restriction on geographical placement

And the following exclusion criteria:

  1. Duplicate results found in another search engine.

  2. Analogous studies reporting similar results, only the most complete study was considered.

  3. Inaccessible studies or books.

  4. Literature that only was available in the form of PowerPoint presentations or abstracts.

Additionally, a fifth exclusion criterion was added in order to tune the search in order to exclude publications that described lake or pond aquaculture or fisheries in such a way that the results would not be applicable to offshore based fisheries and aquaculture:

  1. No studies exclusively describing DSSs for fishery and/or aquaculture in lakes or ponds were considered.

Finally the studies not contributing to answering any of the research questions posed by this our study was excluded.

  1. Only studies contributing to answering our research questions were considered.

No quality assurance or assessment was performed during the search phase, simply to achieve maximum coverage. In agreement with our review protocol, search terms were used for identifying relevant papers in the field of fishery- and aquaculture DSS. We derived our process for synthesizing the query strings in our review protocol from Kitchenham et al. Kitchenham et al. (2007); Kitchenham and Charters (2007). The search terms were selected with trial from a candidate set, which were populated by deriving terms from the research questions. The Boolean “AND” was used to link keywords from different populations in the search strings, the Boolean OR was also used to incorporate alternative spellings. We ended up with the following final search strings:

  • ”Decision support system” AND ”empirical evidence” (Q1)

  • Decision support system fisheries (Q2)

  • Decision support system aquaculture (Q3)

  • ”Decision support system” AND ”fisher*” AND (real-time OR realtime OR real time) (Q4)

  • ”(multi-criteria OR multicriteria OR multi criteria) decision making” AND ”fishery management” (Q5)

It is important to note that we chose to not include “operator support”/“operational support” that can be considered a weak synonym of DSS. However operational support does not share main motivation/stakeholder as DSS and focuses more on operational aspects.

We used EndNote X7.3.1 as our reference manager, and to synthesize our findings, we exported the result set to Microsoft Excel utilizing our separate evaluations to categorize and record details of each paper.

3 Results

The method applied in order to identify relevant studies can be divided into three discrete steps, where the first stage was applying the search query on the data sources. The firs try resulted in 510681 hits. As a result the search queries were refined into the query strings listed in the previous section.

After applying the refined queries across all sources it returned a total of 146 results. More specifically all five query strings where applied to all of the search engines. The five results of from each source was collated, sorted then compared to every other source. The resulting 146 documents where the greatest common denominator of these source. After applying the inclusion- and exclusion criteria, the sum of both manual and automatic searches totaled 70 documents.

3.1 Reading and selective filtering

The data-set containing 70 articles was separately evaluated. Each evaluator was required to compile a list containing all papers complemented by a short description of the paper and its contribution to the research questions this study aims to resolve. Studies not contributing to answering any of the research questions, according to the evaluators, where excluded. The separate evaluations were discussed and compared in order to properly evaluate whether or not the paper should be used in the study. In order to determine the contribution of the articles a majority decision based on the compiled lists had to be reached, to avoid any particular research bias. Juxtaposing the papers resulted in an additional exclusion of 43 papers.

Figure 1 summarizes how we implemented the selection process and presents the number of papers that were identified in each step of the search process. Appendix A presents a complete list of the 27 selected studies, numbered from SID-1 to SID-27.

Figure 1: The different stages of the review. Stage

4 Analysis

This analysis is heavily based on the research questions in the mapping study and is primarily confined into two issues: what decision making systems from fisheries and aquaculture exists, and which empirical results does these systems provide, especially regarding real-time analytics.

4.1 Classification of selected studies and results

We present our findings in the form of a qualitative synthesis, close to what Kitchenham Kitchenham and Charters (2007) describes as a line of argument synthesis as this study tries to infer as much domain knowledge as possible. Consequential to the research questions and future work is the research regarding fishery and aquaculture DSS systems; we therefore start by presenting the number of journals found during the search phase, sorted by publishing year. It should be noted that the numbers presented in Figure 2 only includes the 27 publications that were selected for this study and subsequently forms the basis for answering the research questions.

Figure 2: Publications published on DSS in fishery and aquaculture

While DSS for fisheries and aquaculture have a long history going back to the early 1980’s, e.g. Scuse et al.Scuse and Arnason (1983), the literature surrounding them is sparse. Figure 2 shows several periods during the last 25 years in which the systematic mapping study did not identify any relevant literature published with regards to the criteria previously defined. These periods include the years of 1996, 1997, 2003, 2004, 2010-2012 and 2015. Figure 2 also shows that the topic was at its most popular during the early 2000’s, i.e. 2000-2002. The subject also saw increased popularity in the period of 2007-2009, after which the publication rate fell drastically, as evidenced by the fact that the last five years have seen only one relevant published paper in the form of (SID-27) in 2013. These periods of elevated interest are consistent with the result presented in Arnott et al. Arnott and Pervan (2014), which also notes an increase in publications in these periods.

The declining DSS publishing trend of the last 5 years is not unique to the fishery and aquaculture disciplines as pointed out by  Arnott and Pervan (2014), who notes an overall decline in the number of DSS related publications since the early 1990’s. Arnott et al. Arnott and Pervan (2014) speculates that the decline in DSS publications might stabilize in the coming years as DSS reaches a more balanced position within the domain of information systems, noting that the declining use of DSS might be due to the adoption of other models like the technology acceptance model. While the publishing trend is currently declining, Figure 2 shows that decision support systems within fisheries and aquaculture is still being researched, but to a lesser degree than in previous years.

According to Mardle and Pascoe Mardle and Pascoe (1999)(SID-6), there are few publications on multi-criteria decision making within the field of fishery management compared to other fields like forestry, agriculture, and, finance. SID-6 postulates that in general, the more publications that appear for a given topic, the more research is stimulated, and, thus further publications are generated. The lack of publishing results could then according to Mardle and Pascoe in SID-6 result in an unwillingness to adopt the MCDM technique among the decision makers. Simultaneously, SID-6 conclude that multi-criteria decision making can play an important role in the development of fisheries management policy. The following sections provide a discussion of how each research question was addressed in the mapping study. Results from the mapping study will be presented for each RQ, followed by a discussion of their implications.

4.1.1 (RQ-1) What decision support systems for fisheries and aquaculture exists?

The DSSs found during the review part of the mapping study can coarsely be divided into two categories in accordance with RQ1. The results of this mapping are presented in Table 1. =1.2mm

to 1 X[1,l] X[2,l] Category & Study ID

Fishery & 3, 4, 5, 10, 11, 12, 13, 14, 16, 17, 18, 19, 20, 22, 23, 25, 26, 27

Aquaculture & 1, 2, 6, 7, 8, 9, 14, 15, 21, 24

Table 1: Studies separated into fishery and aquaculture categories by topic.

Table 1 show DSS research to be more heavily focused around fisheries rather than aquaculture; however both fields are well anchored in research and case studies alike.

Noticeably only a single decision support system from Table 1 integrates both fishery and aquaculture requirements in their decision making process, though it should be noted that they are logically separated in the respected system. Except for SID-14 all systems identified in the study focused on either fishery or aquaculture, and the general trend indicates that they should be managed as two distinct classes.

A further classification using the type of DSS described in the papers has been performed, and is shown in Figure 2. Column one of Table 2 shows the classification of DSS within fisheries, while column two shows the same classification for aquaculture. As can be seen from Table 2

, the majority of the fishery DSS systems utilize either a model-driven architecture or neural networks. It seems to be no data driven and significantly few of the remaining types found in this study. This indicates that model-driven- or neural network DSS is the most fitting and applicable methodologies for the fishery DSS domain.

to 1 X[1.4,l] X[1,l] X[1,l] DSS type & Fishery studies & Aquaculture studies

Model driven DSS & 7 & 4

Geographical DSS & 1 & 2

Multi-Criteria DSS & 4 & 4

Grid Based DSS & 1 & 0

Knowledge Based DSS & 1 & 1

Multi-Objective DSS & 1 & 0

Data-Driven DSS & 0 & 1

Neural Network DSS & 3 & 0

Table 2: DSS types grouped by Fishery and Aquaculture.

From Table 2 we can see that model-driven architectures are also popular within aquaculture DSS, however we note that multi-criteria systems are significantly more utilized within aquaculture. Table 2 shows that model-driven systems remain the overall most popular design, but also suggest that the aquaculture and fishery domains contain distinct requirements, which are best solved by differing methods. Also one can observe that there is very little overlap, the sum of types of DSSs in Table 2 is 30 and the number of studies is 27 so only a very few studies actually inhabits more than one type. There is no reason why a DSS using Geographical models cannot also be using neural networks as well. The most likely reasons is research method (focusing on measuring effect of one type of DSS rather than multiple) and effort (implementing more than one is more resources demanding)

4.1.2 (RQ-2) What are the most investigated fishery and aquaculture DSS topics and how have these changed over time.

In addition to classifying the result set based on their type, as seen in Table 

2, the papers were also classified by their problem domains. The results of this classification are presented in Table 3, which shows what papers corresponds to what topic, and Figure 3 in Appendix B which shows how the popularity of the different topics has changed over time. It should be noted that individual papers can correspond to multiple topics in both Table 3 and the table in Appendix B.

to 1 X[1.4,l] X[2,l] Associated DSS topic & Study ID (SID)

Fish health and disease management & 26

Scheduling and planning & 14, 16, 25

Harvest regulations & 4, 13, 19, 22, 25

Sustainability & 4, 11, 13, 17, 19, 22, 25, 27

Catch optimization & 12, 13, 17, 19, 20, 22, 23, 25

Management decisions & 1, 2, 3, 6, 7, 8, 11, 13, 14, 16, 22, 23, 27

Table 3: DSS systems grouped by topics.

Decision support systems for fisheries and aquaculture are very complex tasks as can be seen in e.g. Ernst et al. Ernst et al. (2000), and resolving all parameters influencing decisions which can be mapped to real world scenarios is difficult. This study shows that most fishery- and aquaculture DSS only takes a small subset of criteria into consideration and thus apply only to a single species or a small subset of related species as shown in Table 3. While there are some exceptions to the case where papers cover multiple topics, e.g. SID-13, 19, and 25, it is worth noting that some of the topics covered are often closely related, as is the case with harvest regulations and sustainability. As such, it is natural that papers will often go into related topics as can be seen in Table A in Appendix B, especially when using the categorization of topics that are defined in Table 3. Utilizing the categorization of topics that we have chosen, we found three papers (SID-13, 22 and 25) represented in five different categories, whereas the most common is that a paper is represented in only one or two topics.

Regression analysis performed in Microsoft Excel did not show any noticeable trends regarding the popularity of different topics, and how these topics changed over time, a result most likely due to the small data-set. We reiterate the fact that there are two periods of elevated research in published research i.e. around 2001 and 2009; however these periods are limited in the amount of literature published. This indicates that papers are usually occurring in bulks during a narrow time period, and general purpose DSS as the most recurring theme.

From the table in Appendix B we may infer sustainability and catch optimizations as the most researched form of DSS topics during the most recent years. Considering the increased concern in the most recent years of the global over-fishing problem, it reflects that fishery- and aquaculture DSS might be indicative of political tendencies, especially when coupled with monetary interests. Therefore, we are not surprised to see efficiency and sustainability as popular topics, as these are universal interests that persist across all industries.

Lastly we can see that the analysis of the studies shown in Table 3 shows more overlap than the analysis presented in Table 2. Thus more studies claim to be affecting more topics than they are using different DSS methods/types. Given the hypothesis explaining Table 2 very low degree of overlap - The higher degree of overlap in Table 3 would suggest that it is less resource intensive to test the DSS in another topic and/or that it is less confounding methodologically analyzing the effects of a single DSS type on different topics rather than analyzing different DSS types on one topic.

4.1.3 (RQ-3) Does there exist any DSSs for fisheries providing real-time analytics?

Given the low number of related papers found during the search phase, we found only one paper (SID-19) to fulfill the requirements to be a complete real time fishery DSS, while SID-16, 17 and 23 contains real-time components. We identified sustainability as an important topic in RQ-2 reflecting the growing problem of global over-fishing. As a consequence management decisions should account for sustainable fishing and water pollution in their decision making process. To this end, Caddy and MahonCaddy and Mahon (1995) outlines reference points for fishery management which provides quantitative indicators for system objectives. It states that developing reference points to make quality decisions should be substantiated not only in historical data, but also real time data to best solve their objective. Thus facilitating the usage of real time-data is important for fishery- and aquaculture DSS involving fishing regulations or sustainability. Despite the limited research on this topic, SID-19 shows that real-time data and analytics is possible and has been done previously, although for a narrow area.

4.1.4 (RQ-4) Do DSSes in Fisheries and Aquaculture build their models using machine learning done on captured and grounded data?

An important quality for a DSS is how well it represents the true situation of the decision problem. This quality can be threatened by using out of date models or out of date input data (e.g. simulation or model parameters). In Table 4 we have summarized usage of machine learning and what data the machine learning was trained on. In this context we define machine learning as creating models of phenomenon (e.g. for prediction as in SID-17) automatically based on data. This is in opposition to more static AI methods such as rule based expert systems (e.g. see the traffic light system of Hargrave in SID-15), which is not considered machine learning in this table.

to 1X[l,4] X[l,6] Sub-domain & SID

Fisheries & 12, 16, 17, 20, 25

Aquaculture & 26

Table 4: Studies using machine learning.

We can see from Table 4 that there is only 6 publications that apply any kind of machine learning as a part of their main contribution. More profound in Aquaculture only 1 study (8,33%) uses machine learning. In comparison Fisheries has 5 publications (31,5%) using machine learning as a part of their main contribution.

In Fisheries and Aquaculture data should be available through operational requirements (e.g. video monitoring in aquaculture and catch logging in fisheries). It is an obvious disadvantage for DSS scientists to not draw on these resources for model creation and prediction. This should be a area of focus in future DSS research within this domain.

5 Discussion

This paper presented a systematic mapping study conducted in order to create a solid theoretical foundation for research on DSSs in fisheries and aquaculture by looking at existing research. To our knowledge this is the first mapping study within the field of fishery and aquaculture DSS. The mapping study considered a time period of 25 years from 1990 to the first half of 2015, and resulted in 27 papers that satisfy both the research questions and the inclusion- and exclusion criteria. The results has been validated by the two authors that did not take part in the initial screening (see Section 2.1) that performed a random test to reduce the possibility of any relevant papers being missed when performing the search phase. The validation did not reveal any absence of significant papers, and therefore we believe these 27 papers to accurately describe the current state of the research and work in fishery and aquaculture DSS.

Noticeably throughout the study, very few of the papers found contained rigorous empirical evaluation (see Table 5), however most contained an overview of the system and an outline of their methodology. Some studies used validation data sets, others such as SID-15 actually validated their results against experts and actual decisions. Some studies claims validation but with so short descriptions that the reader is left uncertain with regards to the validation process.

to 1X[l,4] X[l,6] Type of validation & SID

Validated & 15, 19, 26

Validation data-set & 17, 18, 20, 25, 27

Weakly described validation & 12, 21

Table 5: Validation of studies.

Although the resulting set of papers do not provide a unified view of DSS practice, they offer a broad picture and experience on the problem domain. The resulting set of papers could provide those looking to create new DSS within the fishery and aquaculture domains with some helpful insights regarding what methodology is best suited in the design of the DSS depending on the target domain, and an indication of the possibilities within DSSs.

Looking at the research found throughout the mapping study it can be noted that none of the resulting set of papers adequately addresses how to create a decision support system for multiple stakeholders, with multiple criteria with cross cutting concerns for both fisheries and aquaculture. The results of the mapping study provide the reader with several useful and relevant sources of information that could be helpful in the design or research on DSSs in fisheries and aquaculture.

Fishery- and Aquaculture decision support systems aid management as a high level of abstraction on the basis of large quantities of data. Using predictive models, the systems make a qualified guess to increase the probability of an optimal outcome. Designing a system which uses constant up-to-date data for running simulations/applying these data to models poses several difficulties. First and foremost the system is required to keep continuous connections to the various data sources to manage the data; furthermore the system is responsible to complete its analysis within a given deadline. These responsibilities often conflict as no downtime can be expected to complete the deadline responsibilities of the system, and updating data can cause analytics processes to be invalidated during its execution.

We believe most DSS for fisheries and aquaculture does not need to be real-time systems to aid management, and the increased value is often given low priority because of an increased cost, both complexity and monetary.

The mapping study highlights the lack of fishery and aquaculture decision support systems encompassing a multitude of factors for a large geographical area and the species therein. Combining these factors with real-time analytics has not been attempted, most likely due to the complexity of the system (e.g. Ernst et al. (2000)).

5.1 Threats to validity

The primary threats to this study can be identified as whether our approach adequately addresses the principal research questions. The research questions require an assessment of the selected data sources to determine whether we identified all relevant publications and whether our initial classification of the problem domain is denoted correctly for analysis. As there did not exist any previous systematic mapping study on the search terms may miss some terms or synonyms thus missing some publications in the result-set.

The first threat is influenced by both our search criteria, the scope of the search, and the search terms used which are again limited by the search engines capabilities. Our primary method for avoiding these pitfalls has been to employ our third party reviewers to perform a random test on the search queries, the problem domain etc. to find papers we missed in our systematic mapping study. The third party did not find any significant papers missed in this study; therefore we argue that we did manage to provide the most relevant documents published in scientific journals and computer scrience literature. However, unpublished studies have not been considered for this study, and it is therefore possible that we might have missed relevant studies, but overall we presume the study adequately addresses the principal research questions. There exists no former mapping study on this particular domain that the authors could find. Because the topics addressed in this study are selective, there exists little evidence in either direction that we have omitted a major topic that provides substantial empirical evaluation which is in direct relevance to this paper. However it must be noted that we, (the authors), are all software engineers primarily and not researchers in academia, and might therefore be biased in our selection and our analysis, although we have worked to the best of our abilities to avoid this.

6 Conclusion

The finding of our systematic mapping study has implications for those looking to design a multi-criteria decision support system for fisheries and aquaculture. Our findings show that the majority of papers evaluating and documenting their methodology did so with few criteria in mind. The systematic mapping study has identified that very little research has been made into DSS in fisheries and there is a lack of empirical evaluation of these systems. Another concern raised by the study indicates that there exists few to none well documented multi-criteria DSS systems for both fisheries and aquaculture. However the analysis has shown that many components for smaller DSS systems has been well documented, and the result from this study can work as a foundation for further research and development, especially coupled with previous research and knowledge on managing heterogeneous data systems.

7 Acknowledgment

The basis for this work was done in the eSushi project (RCN grant number 245951). Parts of the work were also supported by the EXPOSED SFI project (RCN grant number 237790).


  • Lannan (1993) J. Lannan, User’s Guide to PONDCLASS: Guidelines for Fertilizing Aquaculture Ponds, Pond Dynamics/Aquaculture CRSP, Oregon State University, Corvallis, Oregon .
  • Power (2008) D. J. Power, Decision support systems: a historical overview, in: Handbook on Decision Support Systems 1, Springer, 121–140, 2008.
  • Garg et al. (2005) A. X. Garg, N. K. Adhikari, H. McDonald, M. P. Rosas-Arellano, P. Devereaux, J. Beyene, J. Sam, R. B. Haynes, Effects of computerized clinical decision support systems on practitioner performance and patient outcomes: a systematic review, Jama 293 (10) (2005) 1223–1238.
  • Hunt et al. (1998) D. L. Hunt, R. B. Haynes, S. E. Hanna, K. Smith, Effects of computer-based clinical decision support systems on physician performance and patient outcomes: a systematic review, Jama 280 (15) (1998) 1339–1346.
  • Kawamoto et al. (2005) K. Kawamoto, C. A. Houlihan, E. A. Balas, D. F. Lobach, Improving clinical practice using clinical decision support systems: a systematic review of trials to identify features critical to success, Bmj 330 (7494) (2005) 765.
  • Crossland et al. (1995) M. D. Crossland, B. E. Wynne, W. C. Perkins, Spatial decision support systems: An overview of technology and a test of efficacy, Decision support systems 14 (3) (1995) 219–235.
  • Malczewski (2006) J. Malczewski, GIS-based multicriteria decision analysis: a survey of the literature, International Journal of Geographical Information Science 20 (7) (2006) 703–726.
  • Kitchenham and Charters (2007) B. A. Kitchenham, S. Charters, Guidelines for performing systematic literature reviews in software engineering, Tech. Rep., Keele University, 2007.
  • Budgen et al. (2008) D. Budgen, M. Turner, P. Brereton, B. Kitchenham, Using mapping studies in software engineering, in: Proceedings of PPIG, vol. 8, Lancaster University, 195–204, 2008.
  • Petticrew and Roberts (2008) M. Petticrew, H. Roberts, Systematic reviews in the social sciences: A practical guide, John Wiley & Sons, 2008.
  • Kitchenham et al. (2007)

    B. A. Kitchenham, E. Mendes, G. H. Travassos, Cross versus within-company cost estimation studies: A systematic review, Software Engineering, IEEE Transactions on 33 (5) (2007) 316–329.

  • Scuse and Arnason (1983) D. H. Scuse, A. N. Arnason, INFORMATION MANIPULATION IN BIOLOGICAL DECISION-SUPPORT SYSTEMS, in: Proceedings of the Sixteenth Hawaii International Conference on System Sciences, 1983, vol. 1, Western Periodicals Company, 377, 1983.
  • Arnott and Pervan (2014) D. Arnott, G. Pervan, A critical analysis of decision support systems research revisited: the rise of design science, Journal of Information Technology 29 (4) (2014) 269–293.
  • Mardle and Pascoe (1999) S. Mardle, S. Pascoe, A review of applications of multiple-criteria decision-making techniques to fisheries, Marine Resource Economics (1999) 41–63.
  • Ernst et al. (2000) D. H. Ernst, J. P. Bolte, S. S. Nath, AquaFarm: simulation and decision support for aquaculture facility design and management planning, Aquacultural Engineering 23 (1) (2000) 121–179.
  • Caddy and Mahon (1995) J. F. Caddy, R. Mahon, Reference points for fisheries management, vol. 374, Food and Agriculture Organization of the United Nations Rome, 1995.

Appendix A Selected studies

to 1 — X[1,l] — X[10,l] —

The selected studies.

Study ID (SID) & Reference
(Cont.) The selected studies.
Study ID (SID) & Reference

1 & G. Bourke, F. Stagnitti and B. Mitchell, ”A decision support system for aquaculture research and management”. In: Aquacultural Enginerring(1993)

2 & W. Silvert,”A decision support system for regulating finfish aquaculture”. In: Ecological Modelling(1994)

3 & I. Renaud and S. Yacout, ”Decision support system for quality assurance programs in the fish and seafood processing industry”. In: Computers & Industrial Engineering(1995)

4 & J. Korrûbel, S. Bloomer, K. Cochrane, L. Hutchings and J. Field, ”Forecasting in South African pelagic fisheries management: the use of expert and decision support systems”. In: South African Journal of Marine Science (1998)

5 & S. Mardle and S. Pascoe, ”A review of applications of multiple-criteria decision-making techniques to fisheries”. In: Marine Resource Economics(1999)

6 & J. Bolte, S. Nath and D. Ernst, ”Development of decision support tools for aquaculture: the POND experience”. In: Aquacultural Engineering (2000)

7 & O. F. El-Gayar and P. Leung, ”ADDSS: A tool for regional aquaculture development”. In: Aquacultural Engineering (2000)

8 & D. H. Ernst, J. P. Bolte and S. S .Nath, ”AquaFarm: Simulation and decision support for aquaculture facility design and management planning”. In: Aquacultural Engineering (2000)

9 & S. S. Nath, J. P. Bolte, L. G. Ross and J. Aguilar-Manjarrez, ”Applications of geographical information systems (GIS) for spatial decision support in aquaculture”. In: Aquacultural Engineering(2000)

10 & S. R. Coppola and D. Crosetti, ”Decision-support Systems for Fisheries the ”ITAFISH” Case Study”. In: Studies and Reviews - General Fisheries Commission for the Mediterranean, No. 72 (2001)

11 & K. Hughey, R. Cullen, A. Memon, G. Kerr and N. Wyatt, ”Developing a Decision Support system to manage fisheries externalities in New Zealand’s Exclusive Economic Zone”. In: Modeling and Economic Theory (IIFET 2000) (2001)

12 & S. Mackinson, ”Integrating local and scientific knowledge: an example in fisheries science”. In: Environmental Management (2001)

13 & M. Pan, P. S. Leung and S. G. Pooley, ”A decision support model for fisheries management in Hawaii: a multilevel and multiobjective programming approach”. In: North American Journal of Fisheries Management (2001)

14 & R. Pastres, C. Solidoro, G. Cossarini, D. M. Canu and C. Dejak, ”Managing the rearing of Tapesphilippinarum in the lagoon of Venice: a decision support system”. In: Ecological Modelling (2001)

15 & B. T. Hargrave , ”A traffic light decision system for marine finfish aquaculture siting”. In: Ocean and Coastal Management (2002)

16 & Z. Kemp and G. Meaden, ”Visualization for fisheries management from a spatiotemporal perspective”. In: ICES Journal of Marine Science: Journal duConsell (2002)

17 & A. Iglesias, B. Arcay, A. Rodriguez and M. Cotos,”A support system for fisheries based on neural networks”. In: 1st International Workshop on Artificial Neural Networks and Intelligent Information Processing, ANNIIP 2005- In Conjunction with ICINCO2005 (2005)

18 & T. Koutroumanidis, L. Iliadis and G. K. Sylaios, ”Time-series modeling of fishery landings using ARIMA models and Fuzzy Expected Intervals software”. In: Environmental Modelling & Software (2006)
19 & N. Carrick and B. Ostendorf, ”Development of a spatial Decision Support System (DSS) for the Spencer Gulf penaeid prawn fishery, South Australia”. In: Environmental Modelling & Software (2007)

20 & A. Iglesias, C. Dafonte, B. Arcay and J. M. Cotos, ”Integration of remote sensing techniques and connectionist models for decision support in fishing catches”. In: Environmental Modelling & Software (2007)

21 & R. Wang, D. Chen and Z. Fu, ”AWQEE-DSS: A decision support system for aquaculture water quality evaluation and early-warning”. In: 2006 International Conference on Computation Intelligence and Sequrity ICCIAS 2006 (2007)

22 & F. Azadivar, T. Truong and Y. Jiao, ”A decision support system for fisheries management using operations research and systems science approach”. In: Expert Systems with Applications (2009)

23 & R. V. Chandran, A. Jeyaram, V. Jayaraman, S. Manoj, K. Rajitha and C. K. Mukherjee, ”Prioritization of satellite-derived potential fishery grounds: An analytical hierarchical approach-based model using spatial and non-spatial data”. In: International Journal of Remote Sensing (2009)

24 & H. Halide, A. Stigebrandt, M. Rehbein and A. McKinnon, ”Developing a decision support system for sustainable cage aquaculture”. In: Environmental Modelling & Software (2009)

25 & L. Sun, H. Xiao, D. Yang and S. Li, ”Intelligent decision support system for fisheries management”. In: Journal of Computational Information Systems (2009)

26 & Z. Xiaoshuan, F. Zetian, C. Wengui, T. Dong and Z. Jian, ”Applying evolutionary prototyping model in developing FIDSS: An intelligent decision support system for fish disease/health management”. In: Expert Systems with Applications (2009)

27 & Y. K.Teniwut and Marimin, ”Decision support system for increasing sustainable productivity on fishery agroindustry supply chain”. In: 5th International Conference on Advanced Computer Science and Information Systems, ICACSIS (2013)

Appendix B Timeline of DSS topics

Figure 3: Timeline of DSS topics in selected studies