--- title: I rant about Dars, while Marko Plahuta gathered the data and showed what is really going on date: 2026-10-06T15:25:08Z modified: 2026-10-06T15:29:29Z permalink: "https://www.had.si/blog/en/2026/10/06/i-rant-about-dars-while-marko-plahuta-gathered-the-data-and-showed-what-is-really-going-on/" type: post status: publish excerpt: "Marko Plahuta's analysis of Slovenian motorway data shows that recurring bottlenecks, roadworks and poor traffic management are behind far more than just summer congestion." wpid: 390478 categories: - Reflections tags: - Reflections - dars - Kozarje - Marko Plahuta - motorways - roadworks - traffic accidents - traffic management - Virostatiq - english - pll_6ac51348c0813 language: - english post_translations: - pll_6ac51348c0813 featured_image: "https://www.had.si/blog/wp-content/uploads/2026/10/virostatiq_dars_marko_plahuta.jpg" featured_image_alt: virostatiq dars marko plahuta timestamp: 2026-10-06T15:29:29Z --- [Marko Plahuta aka virostatiq](https://virostatiq.com/), is not a traffic expert and does not present himself as one. He is a software engineer, a data-mining specialist and someone who clearly enjoys tackling problems that most people do not even notice until you start looking at them through data. That is precisely what makes his work so interesting. He takes vast quantities of apparently unrelated data, brings them together, analyses them and tries to extract patterns that are difficult to spot with the naked eye. He turns problems we feel must exist into something measurable and visible. - [How much do traffic jams cost Slovenia? More than €2 million a day](https://www.had.si/blog/wp-content/uploads/wp-mfa-exports/post/how-much-do-traffic-jams-cost-slovenia.md) - [Tesla After 65,000 km: I No Longer Think About Driving an Electric Car. I Just Drive a Car](https://www.had.si/blog/wp-content/uploads/wp-mfa-exports/post/tesla-after-65000-km-ownership-costs.md) - [AI probably won’t take your job. Someone who knows how to use AI better than you might](https://www.had.si/blog/wp-content/uploads/wp-mfa-exports/post/ai-probably-wont-take-your-job-someone-who-knows-how-to-use-ai-better-than-you-might.md) This time, he turned his attention to Slovenian motorways. On [X, he published a long thread on the state and traffic flow of our motorway network](https://x.com/Virostatiq/status/2107380213864673283), based not on a few photos of queues or one bad Friday drive to the coast. Since April, he has been collecting data from publicly available traffic counters at five-minute intervals, then linking it with congestion, accidents, roadworks and police data. He also detected malfunctioning counters and interruptions in data sources, and factored them into his analysis. This is where it became interesting to me as well. Because the data demonstrate much of what I have been saying for a long time, based on my own experience of Slovenian motorways. The problem is not just tourists, it is not merely the number of cars, and the answer to every traffic problem is not necessarily another lane of tarmac. A large part of the problem lies in how the network is managed, when works are carried out, how roadworks are organised and what happens when you remove part of the capacity from a system that is already under strain. When you look at it all together, a rather unpleasant picture emerges. Traffic jams are not random events that happen here one time and somewhere else the next. They recur in very specific places, at very specific times and on very specific stretches of road. Ravbarkomanda is one such example, but the Kozarje interchange is even more interesting. Virostatiq finds that two major traffic flows merge there, that the problem behaves in a markedly non-linear way, and that it then affects a far wider part of the network. He even writes that, if he could make only one intervention, he would address the convergence at Kozarje in the direction of Primorska. Even more interesting is his finding about so-called traffic hysteresis. Traffic flow can collapse very quickly, then take considerably longer to recover. That means a disruption lasting only a few minutes, in the wrong place and at the wrong time, does not mean a delay of just a few minutes. It can trigger a chain reaction that drags on long after the original problem has disappeared. Then we get to the part where the analysis becomes rather uncomfortable for Dars. Virostatiq does not infer from the data that every traffic jam is caused by roadworks, or that heavier traffic plays no role. On the contrary, he clearly says that increased traffic matters at particular peaks and in particular places. But he also finds that roadworks and the disorder associated with them are a very important part of the problem, and that the explanation that there are simply too many cars on the roads does not explain enough. That is exactly what has bothered me about Dars for a long time. Of course there is more traffic in summer. Of course Fridays are a problem. Of course people come into Ljubljana in the morning and leave it in the afternoon. Of course tourists increase traffic towards the coast in July and August. But none of this is a surprise. Dars knows when Friday is. It knows when August is. It knows when the morning and afternoon peaks are. These are all things you can predict far more accurately than the weather two weeks from now. This is not a natural disaster. It is a calendar. If you manage a system, how you plan interventions in that system matters a great deal. The question is not whether roads need to be renovated, because of course they do. The question is whether you organise works in a way that causes the least possible damage to traffic flow, or whether you put up a closure, reduce capacity and then tell users that the problem is simply too many cars. The most uncomfortable part of virostatiq’s analysis, though, concerns accidents. According to his data, the likelihood of an accident is around 3.6 times higher in a traffic jam than in free-flowing traffic. Rear-end collisions predominate, but side-on collisions and scrapes also occur, which is a fairly logical outcome of slow traffic, lane changes and constant starting and braking. It is also interesting that congestion affects accident risk even more strongly than heavy rain. That obviously does not mean every traffic jam is a direct cause of an accident, or that we are dealing with some perfect causal relationship. But it does mean the connection is large enough that a road operator simply cannot ignore it. When virostatiq compares this year’s situation with data from previous years, he shows an increase in the number of accidents on several stretches where works were under way this year. He is considerably harsher in his own conclusions, speaking of very poor management and the responsibility of the people running such a system. These things raise a perfectly legitimate question: are roadworks and the traffic arrangements around them designed safely enough and efficiently enough? Above all, are they based on expert decisions? Because if you know that congestion substantially increases the risk of accidents, and if you see the number of accidents rising on stretches with roadworks, it is not enough simply to put up a 60 sign and say drivers are responsible for everything. The system itself must be designed with real drivers in mind, not ideal drivers from a manual. Another point from his analysis is very interesting, and one we often do not even realise. When a motorway becomes clogged, the traffic does not disappear. Drivers start leaving the motorway and moving onto local and regional roads. That means motorway congestion is no longer just a problem for someone sitting in a car on the motorway. It becomes a problem for people living in the places alongside it. Virostatiq’s data show that accidents also increase on the roads where traffic spills over when the motorway is congested. This matters because it clearly shows that poor management of the main transport artery does not have consequences only on that artery, but spreads them across the entire surrounding area. Then we come to tourists, who are the convenient answer here to almost every summer traffic jam. Virostatiq does not exclude tourists from the equation at all. Quite the opposite: he writes that they undoubtedly contribute to both congestion and accidents. But the essential question is not whether tourists increase traffic, because of course they do. The essential question is how a system can act surprised every single year by their arrival. Tourists do not turn up unannounced. They do not appear on some random Wednesday in November. Their arrival can be predicted, traffic peaks can be measured, loads can be simulated, and problematic stretches can be identified years in advance. If you have data, you also have the ability to manage. That is what I find most important about virostatiq’s analysis. Not because it proves Dars is to blame for every traffic jam, every accident and every lost day of someone’s holiday. It does not prove that, and such a conclusion would be too simplistic. But it does show that problems recur in the same places, that some bottlenecks are particularly troublesome, that roadworks have a major impact on traffic flow, that congestion is associated with a substantially higher accident risk, and that the consequences spill from the motorway onto local roads as well. Finally, we come to Slovenia’s eternal answer to traffic problems: a third lane. Perhaps we really do need one on certain stretches. We probably need one in many places. But a third lane will not, by itself, fix poor planning of works, badly organised merging, problematic interchanges and a system that collapses at the first serious disruption faster than it can recover afterwards. If the way it is managed stays the same, one day we will simply be sitting still in three lanes instead of two. That is what I have been saying for a long time. Slovenia does not only have a problem with too few roads or too many cars. It also has a traffic management problem. It has a problem with decision-makers who have clearly never carried out the kind of analysis Marko Plahuta has. The greatest irony is that we no longer need to rely on feelings and pub debates. The data exist. Traffic is measured, congestion is measured, accidents are measured, we know where the system collapses, and we also know when it happens most often. If Marko Plahuta can gather public data for several months and use them to build such a picture of Slovenia’s motorway system, then I would expect Dars to have considerably more data available, considerably better models and considerably more people working on this. The only question is what they do with them. Because perhaps the problem with Slovenian motorways is not just that we need more tarmac. Perhaps we should first manage the tarmac we already have much better. Or perhaps Dars is simply a parking place for certain people who do not give a damn about safety, congestion and, above all, the state of our motorways. Despite the fact that we lose more than €800,000,000 in congestion every year! Marko, thank you! And I hope the people at Dars call him, so he can explain the problems to them and offer solutions based on concrete data they are not even aware of themselves.