Ingår som bilaga till förf:s licentiatuppsats: "Prognos av vägars bärförmåga vid tjällossningen: användning av temperatur som nyckeltal" Luleå, 2010

Många vägar skadas varje år av tjäle och dessa vägar behöver åtgärdas. Åtgärdsbehoven är i regel större än resurserna och därför måste en insatsprioritering ske. En korrekt utförd tjälinventering ligger till grund för en korrekt åtgärdsplanering. Tjälinventering genomförs idag av fysiska personer och personerna kan variera från år till år. De företag som genomför inventeringarna kan även variera beroende på offert, kapacitet, kunskap, erfarenhet etc. Att olika företag och personer genomför inventeringen kan medföra systematiska skillnader. En väg kan alltså tjälinventeras av olika aktörer år till år beroende på vilka som får uppdraget. Vid en förstudie av tjälinventeringar från ett år (2008) gjordes jämförelser med följande års (2009) inventeringar varvid man fann stora skillnader i resultat trots att det var samma vägsträcka som studerades. Eftersom de två jämförda inventeringarna inte visade likvärdiga resultat har frågan om väginventeringens kvalité väckts. Ett projekt har initierats av Trafikverket för att studera inventeringsförfarandet som sådant samt undersöka skillnaderna mellan inventeringarna och skillnadernas omfattning. Metoden som valdes för studien bestod av att fyra olika aktörer, Konsult A respektive B, Trafikverket (TV) samt personer från Luleå tekniska universitet (LTU) fick inventera en och samma vägsträcka vid samma dag och tidpunkt. De olika aktörerna hade olika stor erfarenhet av inventering och den sträckte sig från enbart teoretisk kunskap till mångårig praktisk sådan. Ordnade efter praktisk erfarenhet av inventering har Konsult B mest erfarenhet, därefter Konsult A och TV. Minst erfarenhet hade LTU. Vägsträckan som inventerades var väg AC805 från Lidbacken till Femvägaskälet utanför Skellefteå i Västerbottens län. Vägsträckans totala längd var cirka 15 km och vägens ÅDT var 576. Med hjälp av bedömningarna valdes fyra delsträckor ut för vidare studium. Dessa delsträckor representerade vägsträckan i stort och visade på skillnader och likheter mellan de olika inventerarna.

Roads are affected by frost. For maintenance and action planning of these roads, a well performed inventory is a necessary base. It will also be a base for priority of different objects for repair and maintenance. Therefore, it is very important to have a similar ground for the judgement. This can be difficult to achieve as different companies and persons are enrolled different years due to type of contract used, capacity, experience, knowledge etc. In a prestudy great differences were found when comparing two inventories done on the same road in year 2008 and 2009. The question was then raised, if the provided information on how to perform an inventory was good enough in order to carry out a high quality work. Another question raised, was if the differences really are that obvious between different persons performing the inventory. The method used for the study was to have four actors to carry out an inventory each on the same road, at the same time, and then compare the obtained results. The actors were “Konsult A” and B, Swedish Transport Association (TV) and persons from Luleå University of Technology (LTU). Ranked in order of practical experience the most experienced was Konsult B, then Konsult A, TV and LTU. The road studied was AC805 Lidbacken to Femvägaskälet. It is situated outside Skellefteå in the county of Västerbotten in the north part of Sweden. The total length of the studied road was about 15 km and it has an ADT of 576. Four sections, which represented the road as a whole and also showed the main differences and similarities, were closer analyzed. One section focused on an overview, one on cracks and two sections focused on bumps and roughness. In the report, the results from the four inventories are presented graphically in order to give the reader an overview and a chance to comparison. The results show that there are both great differences and similarities between the four inventories. The two most experienced, Konsult A and Konsult B had about the same percentage of their total crack length in class 2 (66 respectively 60%) and class 3 (~30%). The total amount of inventoried crack length differs however. Konsult A have in total 7355m and Konsult B 9157m, which means that Konsult A only reports approximately 80% of the length reported by Konsult B. TV had 18% class 2 and 82% class 3 cracks with a total length of cracks being 3246m, while LTU had 38% class 2 and 24% class 3 cracks. In this case the total crack length was reported to be 11029m. This means, that the total length of reported cracks varies between 3246 m to 11029 m, corresponding to approx. 21% and 74% of the total road length. The most experienced (Konsult B) and the least experienced performer (LTU) show most similarities in the overall road condition assessment and placement of damages. It is mentally exhausting to perform a road inventory by hand without small amount of technical support. It is shown in the report that the resolution seems to decrease after some time of work. As a consequence, many cracks in class 1 are registered in the beginning of the studied road stretch, while in the end cracks are more registered as long single cracks. After some time of work more cracks are registered as coherent cracks while they in fact are single cracks. It should therefore be considered if technical supporting systems, like e.g. GPS and computers with well developed software, should be more used in order to increase efficiency and quality of the final work. This should also reduce the impact of subjective decisions. One observation done in the study is that different persons might have different views of the purpose of the road inventory. One view is to perform just a road inventory, without any specific purpose, more than to get a view of the quality of the road. Another point of view is to do an inventory with the focus of what type of actions are to be carried out in future in mind. Depending on the focus, the results seem to differ. In this study, it seems like LTU, Konsult A and Konsult B have performed the inventory in the first way, while TV seems to have carried the inventory out with future actions in mind. This conclusion is based upon the observation that TV did not note any damages in class 1 throughout the whole road stretch, while the other did. By analysing the inventory results, it was found that errors in the distance measurements were present. It is therefore proposed that the customer should consider GPS-length measurements to eliminate this source of error. Inventory done with the base of digital photos will minimize some of the length measurement problems if the photos are taken by a camera connected to a GPS. However, inventory done by using digital photos only have several disadvantages in comparison with ordinary field inventory. The major drawback of this methodology is the difficulty to discover small damages and roughness. It is difficult to classify roughness. The roughness effect on driving experience is very much dependent on the speed of the inventory vehicle. Based on this study the speed of the inventory vehicle needs to be considered when doing a roughness inventory. Roughness inventory should be done at the roads intended speed to ensure that the roughness impact on driving comfort is correctly experienced and registered. One way to enhance the quality of road inventory could be to develop a clear and easy to understand description of method. It should be emphasised that experience is a key component when doing road a inventory, but an increased use of technical support systems can reduce the scatter in the assessment.