Journal editors like to believe they are more than mere traffic cops. But here’s a case that makes us wonder.
Transportation Research Part E: Logistics and Transportation Review — yes, there are parts A-D of this — is retracting a 2009 paper which stole liberally from a 1996 article in a different journal from the same publisher, in this case Elsevier.
According to the notice, which appeared online earlier this year and in the July 2011 issue:
The authors have duplicated significant parts of a paper published in Fuzzy Set. Syst., 82 (1996) 307–317, doi:10.1016/0165-0114(95)00276-6. One of the conditions of submission of a paper for publication is that authors declare explicitly that their work is original and has not appeared in a publication elsewhere. Re-use of any data should be appropriately cited. As such this article represents a severe abuse of the scientific publishing system. The scientific community takes a very strong view on this matter and we apologize to readers of the journal that this was not detected during the submission process.
‘‘The Vehicle Routing Problem with Uncertain Demand at Nodes’’, plagiarised the paper ‘‘The Fuzzy Set Theory Approach to the Vehicle Routing Problem when Demand at Nodes is Uncertain’’ by Teodorovic and Pavkovic that appeared in 1996 in the journal Fuzzy Sets and Systems.
Here’s the first line from the retracted article’s abstract:
The vehicle routing problem with uncertain demand at nodes is considered.
Here’s the first line from the abstract from the original paper in Fuzzy Sets and Systems (not to be confused with the Journal of Fuzzy Dice):
The problem of vehicle routing when demand at the nodes is uncertain is considered.
The plagiarists, from Hunan Province in China, then lay out their “Problem assumptions”:
We assume that there are n nodes in the network to be served. Vehicles set out from depot C, serve a number of nodes and upon completion of their service, return to the depot (Fig. 1). We also assume that the demand at each node is only approximately known. Such demand can be represented by a triangular fuzzy number (Fig. 2). Fig. 2 presents the membership function of triangular fuzzy number D representing demand at the node.
But what do we have here? Why, “Problem assumptions” from Teodorovic and Pavkovic, who are from Belgrade (now in Serbia but then Yugoslavia):
Let us assume that there are n nodes in the network to be served. Vehicles set out from depot [D], serve a number of nodes and upon completion of their service, return to the depot (Fig. 1). We also assume that the demand at each node is only approximately known. Such demand can be represented by a triangular fuzzy number (Fig. 2). Fig. 2 presents the membership function of triangular fuzzy number O representing demand at the node.
And what about those figures?
Figure 1 from the Chinese authors:
And from the Yugoslav researchers:
Figure 2 from the plagiarists:
And Figure 2 from the plagiarized:
Now, obviously a lot of this stuff is easier to detect in hindsight, and we’re not suggesting that editors can catch everything. But until journals get more serious about their monitoring efforts, it’s not going to improve. After all, this isn’t a “community problem,” it’s a high-tech battle against fraud. And although it will pain Elsevier to hear it, the fact that it can happen within the same publisher suggests that the flow of information within the company and across titles isn’t what it could be.
The problem with putting the onus on the editors and reviewers is simply the vast amount of stuff out there, as well as the obscure nature of many journals. Additionally, there is the current “horizon” of full-text publication. I would just guess that many obscure pieces might be subject to plagarization if they were published a) in obscure journals b) before 1995.
I disagree to some extent here. It really is not difficult to run the keywords through a search engine, and in the case of Elsevier reviewers get access to Scopus which shows similar papers (even ranks them according to similarity). The plagiarised paper should have shown up right at the top.
If the earlier publication is available in full text on-line and the new submission is run through one of the plagiarism detection websites, it might be detected. Otherwise, there is no way that a journal editor could possibly detect this type of plagiarism. The fact that the two journals are from the same publisher is irrelevant.
Disclosure: I am associate editor of a medical journal.
I agree that you can’t just blame Elsevier here, and perhaps not even the journal editors. The reviewers failed, obviously, as the Elsevier review system gives you access to Scopus, which highlights relevant similar articles. I always look at that list to see if something similar was published before. The reviewers obviously did not.
I think one issue with a lot of plagiarism software, even if it WAS used here, is that the original paper is available online, but as a scanned copy of the paper version. Does anyone know whether plagiarism software (and which) can handle scanned copies versus ‘text’ pdf’s?