Overview of German Additive Manufacturing Companies

This dataset is the description of a curated list of companies involved in additive manufacturing in Germany. The companies included are of various categories, such as 3D printing providers, hardware manufacturers, software developers and vendors. The list was compiled through literature and Internet-based research, resulting in the compilation of information from a number of resources, such as the Bundesanzeiger (Federal Gazette), the Registergerichte (Register Courts), the respective websites themselves and a B2B marketplace (Wer liefert Was?). The aim of compiling this list is to provide information to researchers on the current situation of 3D printing in Germany. Data Set: available as the supplementary files Data Set License: CC-BY-NC-SA


Introduction
3D printing and additive manufacturing (AM) are techniques and technologies to manufacture or create physical objects out of a number of materials such as plastics, metals or ceramics from digital model files generatively without requiring special tools besides a 3D printer.This technology is currently highly researched, and developments have been made by public institutions and private entities.Data on the prevalence of certain technologies in general or the market value in individual countries is sparse.This dataset is compiled to provide a basis for further research, such as market analysis and capitalization analysis, in Germany.The list contains information about 284 companies, ranging from individual persons, acting as legal entities, to internationally-active corporations.Information on such entities can also be used to incorporate external services in cloud manufacturing systems or connected 3D printing services, as described in Baumann et al. [1].Furthermore, this information can be used to help to identify stakeholders in resource description endeavors as described in Baumann and Roller [2].This dataset enables research on company structure and historical development in the domain of AM.Furthermore, with this dataset, collaboration initiation and the opportunity for industrial interested parties are facilitated.Functioning as a basis for extension, the dataset can also be used to assess the impact of this technology on the national economy of Germany.
This work is structured as follows: Following Sections 1 and 2 provides information on the method and sources used in this research for the data acquisition and compilation.Section 3 provides information on various sources used for the generation of this overview, on the motivation for the respective source selection and provides information on the scripts employed for automatic data acquisition in Section 3.2.Section 4 contains information on the dataset provided with this research as supplemental data, and Section 4.1 provides detailed information on the structure of the overview document containing all of the information on the companies, excluding the downloaded html files.

Data Acquisition
The compilation was performed via literature (see Fastermann [3]) and Internet-based research for certain keywords and phrases.As sources for the Internet research, the search engines Google [4] and Bing [5] were utilized.The search phrases include, as well as combinations thereof:

•
+"3D Druck" + Firma (engl.translation:3D print + company) The search phrases and keywords are presented here in their original German language, as the research is focused on Germany and the justification is that the companies sought address the German market in German.It was also argued that searching in this language will eliminate a large number of businesses that are out of scope, i.e., from other countries than Germany.
Furthermore, information was acquired indirectly via websites of federations or unions on AM-related topics.
For each company found, the respective website was visited, and the information of this site was extracted both manually and automatically, with the support of custom scripts (see Section 3.2).The German law dictates that every business operating in Germany and having a website must provide certain information in their imprint, such as the name, address and routing information.International companies with only German subsidiaries, such as Makerbot Deutschland, were not considered for this compilation.Furthermore, companies with only limited involvement in AM, such as Verbatim, who sells 3D printer filament, where apparent, were also excluded from consideration.Companies listed in association information, as described below, with apparent limited involvement in AM, such as companies offering photography services for 3D printing projects, were also excluded.
The list of excluded companies from the neighboring European and overseas American companies include: These companies are either subsidiaries of foreign companies, operate in Germany from abroad or are located in neighboring countries and are of similar legal standing as German companies, e.g., a GmbH, Gesellschaft mit beschränkter Haftung, engl.company with limited liability.
Companies were further identified from the following associations, unions and exhibition websites:

Results
For the compilation of the information on the companies, information from the following online sources was used.The relevance and information acquired is described after the list.

•
Bundesanzeiger/Federal Gazette [6]  Handelsregister/Registry court [12] https://www.handelsregister.deInformation on the company was acquired primarily from the company website, which was identified by the means described herein.The Bundesanzeiger website was used to retrieve the latest balance sheets for the company where available.The latest balance sheets available were from the fiscal year 2015 for most companies.The information from Archive.org was used to assess the first and last archival dates for the website to estimate the popularity of the website (more popular websites are archived more often than less popular websites) and estimate the first appearance of the website on the Internet, which might coincide with the inception date of the company.Xing was used to acquire information on the company owner and might be used in later research to assess the company structure and employee qualification.Information from Kununu was used to assess a rating on the company from the viewpoint of its current and former employees and might be used in later research for qualitative dissemination.Geocode was used to map location information such as city to geo-information for display in digital maps; see Figure 1 for an example.The B2B marketplace Wer liefert was? (WLW) was used to provide additional information on the company such as inception date, categorization, description and contact information.Information on this website is provided by the company itself.The search engine Bing was used to provide information on the number of entries in the cache from the search engine with the motivation to assess the popularity of the website, with more popular websites having more entries in the search engine cache.Information from the registry courts was used to provide official information on the company address and status, as well as initiation documents, which are provided separately.
After probing fifteen randomly-selected companies from the dataset without a single result, it was deemed that the LEI (legal entity identifier) from the Legal Entity Register, https://www.leireg.de, is insufficient to provide meaningful information to the data in its current state.

Company Name
• Sale or production of hardware, i.e., 3D printers (indicated as HW(P), in the overview document) • Sale or development of software, e.g., 3D printing simulation or other tools (indicated as SW, in the overview document) • Sale of supplementary material for 3D printers, e.g., filament, powder, or resin (indicated as HW(MAT), in the overview document); also development of 3D printing material • 3D printing service and other services, such as 3D printer rental, 3D model preparation or training (indicated as SERVICE, in the overview document)

Script-Based Information Retrieval
To facilitate efficient data acquisition and enrichment, a number of scripts were employed.The scripts are written in BASH [13] for the Linux platform.These scripts are detailed in the following Sections 3.2.1-3.2.7.

Registry Court Information
See Supplementary Material reg-ger-search.shfor the script that was used to acquire data from the central registry court information website.This script scrapes the website for the search term that is the company's name and then downloads all associated publications and information for the respective company.A number of publications, such as historical records, must be purchased from the registry court website, and these publications are excluded from the download.The results were manually checked to eliminate false-positive results for companies with similar names.

Information from B2B Marketplace WLW
The script wlw-search.shscrapes the B2B-marketplace Wer liefert was? (WLW) for information on the company in question.The script utilizes the central search function of the website and acquires the details present for the first company by the searched for name.Furthermore, this script extracts the html title of the associated company website to provide an automated way of capturing this information.The WLW website contains information provided by the businesses themselves.On this website, all registered companies are identified by an integer identifier.

Information on the Results from the Search Engine Bing
With the script bing-info.sh,information on the number of results in the search engine Bing's cache is extracted, as well as information on how often the company URL is linked on other websites.

Information from the Internet Archive
The script archive-info.shparses the information from the Internet Archive at https://www.archive.orgfor a specific URL, the company URL.Information is extracted on the first and last occurrence of the archival process, with the motivation that popular websites are being archived more often and, furthermore, to provide information on a possible company inception date.Information on the number of archival requests is also retrieved.

Wordcloud Generation from Website Data
The script website2wordcloud.shextracts all visible and invisible links from the company website's base URL and downloads every document linked to a temporary folder.From the downloaded files, only files that are html are further analyzed.The resulting files are parsed from html to text and aggregated in one single file for wordcloud processing.The wordcloud processing is performed with the software word_cloud [14] from https://github.com/amueller/word_cloud,and the most common words are placed on a digital canvas according to their occurrence and scaled accordingly.For further processing, the aggregated text file is separated into words and stored separately.An example for such a wordcloud is presented in Figure 2 from the company StaeGi GmbH with the URL http://www.staegi.de.A number of websites was not suitable for automatic parsing and acquisition due to heavy reliance on JavaScript, Flash or other technology that is not readily accessible with text-based tools.These websites are unfriendly to screen readers and other assistive technology and were excluded from further automated analysis.

Most Common Words on Website
The script most-common-words.shanalyzes the aggregated textual website data generated by the website2wordcloud.shscript for the ten most used words.The words are compared against a list

Structure of Data
The following list explains the data structure that is present in the company list.One example each, description or name, is provided for every column, starting from the first:

Figure 3 .
Figure 3. Wordcloud on all downloaded and aggregated websites.