Working with 1Spatial, AdV’s ATKIS-Gen Project Group’s cartographic output is 40% faster with automated generalization.
The use of location information by business and consumer continues to rise and organisations like national mapping agencies are facing increasing demand for their authoritative geospatial data.
For Germany’s mapping agencies, producing a single 1:50,000 map sheet took three months. The state of Rheinland-Pfalz alone is represented by 41 sheets at this scale, and required two cartographers working full-time to maintain its five-year production cycle.
In Germany, each of the country’s 16 Länder or states is responsible for surveying and mapping its territory. The AdV (short for Arbeitsgemeinschaft der Vermessungsverwaltungen der Länder der Bundesrepublik Deutschland) is the committee responsible for coordinating activities. It acts as a single point of contact for customers and coordinates strategy, standards and processes for the states.
Historically, the AdV maintained three separate types of spatial data, each with its own information system. These were: cadastral (land registry information – ALKIS), topographic/cartographic (mapping data – ATKIS) and geodetic (data pertaining to trigonometric points and mapping projections – AFIS).
However, recognising the redundancies involved and the potential for conflict where features were handled differently in each system, the organisation developed AAA. This is a common data model that includes all data collected by the 16 state surveyors.
The combined basis data is enormous (for the state of Rheinland-Pfalz alone, ALKIS contains 177 million separate geospatial features). As Dr. Dietrich Schürer, Head of the AdV’s ATKIS-Generalization Technical Committee, describes, “In AAA, we developed one base model, aligned to OGC international standards, where information can be captured and stored once, but used many times by different systems and customers.”
Towards customer-centred efficiency
The AdV provides geospatial information to other federal agencies (such as the Federal Environment Agency, Federal Police and border control) and to private companies like Google, Deutsche Bahn AG (the German railway company) and Garmin.
In the past, the AdV had worked to a five-year production cycle for all of its map products. However, customers increasingly required information to be updated more frequently. Many also wanted specialised maps, tailored to the exact needs of their digital systems. These requests were impossible to meet with the surveying authorities’ existing resources.
Twelve of the 16 AdV member states decided to develop an automatic process to speed up production and reduce manual effort. They founded the AdV ATKIS-Gen project group, because it would not be possible to meet increased demand for its services without either reducing the range of maps and scales that it produced, or investing in making the production process more efficient. Reducing the quality of the products was not an option.
Innovation through automatic generalization
The AdV ATKIS-Gen project group evaluated potential partners from different fields to help make production more efficient. However, Dr. Schürer explains, “When we looked at universities who were researching cartographic generalization, they were only focusing on particular problems whereas we wanted a holistic approach. It was clear that we needed a strong partner with deep knowledge and experience of handling geospatial data.”
1Spatial was selected as the AdV ATKIS-Gen project group’s partner and together they developed an automated, “context-aware” solution using 1Spatial’s experience and software tools. The innovative project was unique in its holistic approach and in its scale. It resulted in one of the largest generalization projects in Europe.
A goal was set to reduce the production cycle to just three years by finding ways to automate the generalization process. The AdV ATKIS-Gen project group’s future target was to enable Just in Time, or On Demand, map production.
40% improvement in production time
With the data harmonised, inconsistencies resolved and the solution being used across the states, the production cycle became 40% faster, reducing from five to just three years as planned.
The approach, which consists of model generalization and cartographic generalization, is now used for the production of all maps.
Basic model generalization for a whole state can now be achieved in just one or two days and the AdV is on track to meet its future target of Just in Time production.
Meeting customers’ complex requirements
“Auto-generalization ensures that all states can deliver data in the same structure and to the same quality for the customer. Our work is enhanced by the digital workflow,” explains Dr. Schürer.
The AdV now offers a web service for customers (WebAtlasDE), providing consistent, authoritative, geospatial data for the whole of Germany.
Despite reducing staff numbers, the AdV ATKIS-Gen project group can also respond more quickly to requests from customers with complex needs. Custom maps can be prepared to special requirements and, if necessary, the AdV ATKIS-Gen project group staff can provide consultancy; helping to import data into customers’ systems or producing customised datasets that combine the AdV ATKIS-Gen project group and customer data.
Not only is the AdV ATKIS-Gen project group now more efficient overall, the work of individual cartographers has become more rewarding. Routine tasks are completed within an automatically generalized map and the experts can focus their skills and knowledge on the remaining, more exacting problems.
“We need good cartographers with great experience to deal with the most difficult problems, because the smaller issues are solved automatically,” Dr. Schürer explains.
Increasingly, AdV sees its role as a central source of reliable, authoritative geospatial data that supports the continuing development of society and the economy. As Dr. Schürer notes, “We produce data for all people, the public as well as business, and we work very hard to bring good data to everyone. After all, we believe that 80% of every decision is based on geospatial data.”
As NMAs and other organisations use approaches like repeatable, rules-based automation to drive efficiency, they are better placed to meet growing demand for their services and the authoritative geospatial data that they manage.
Both 1Spatial (www.1spatial.com) and IP-ATKIS-Gen will be presenting on generalization at the 28th International Cartographic Conference in Washington, DC, July 2 to 7, 2017. Visit the 1Spatial stand (booth 418) to find out more about overcoming data challenges and getting data ready for generalization.