University Patent Activity in Saxony & Thuringia

The regional academic portfolio is one large filer plus a long tail: TU Dresden holds almost half of all families, while eight of the seventeen institutions hold fewer than 30 each. The data also shows a portfolio that has become less national over time — the EP and Chinese shares both rose between the two complete five-year windows while the DPMA share fell — and one that leans on a single external partner, Fraunhofer, for 11.4% of all families.

TU Dresden holds 47.0% of all regional academic patent families (1,032 of 2,197) and 47.2% of all applications. The second tier — TU Bergakademie Freiberg (236), FSU Jena (224) and TU Ilmenau (209) — is close together, and the gap to TU Dresden is more than a factor of four. Eight institutions hold fewer than 30 families each over twelve years.

Filing routes differ sharply by institution. FSU Jena reaches abroad most consistently: 152 of its 224 families (67.9%) contain a WO member and 92 (41.1%) have an EP filing. HS Mittweida sits at the other end — 69 of its 71 families have a German filing and only 4 contain a WO member. Universitätsklinikum Jena is the only institution whose EP filings (42) outnumber its German ones (23).

Regional academic filing volume was at its highest at the start of the period and has not returned to that level. 2013 to 2017 saw filings fall from 281 to 163 families — a drop of 42.0% over four years — after which the series moves within a narrow band of 156 to 185 families per year through 2023. Excluding the incomplete 2024, the last five complete years (2019–2023) average 168.8 families against 223.4 for 2013–2017.

The decline is not evenly shared. Thuringia fell from 78 families in 2013 to 31 in 2019 and has stayed between 31 and 45 since; Saxony fell from 203 to 109 by 2017 but recovered to 122–145 in 2019–2023. TU Dresden shows the same pattern within Saxony — 135 families in 2013, a low of 67 in 2018, then 72–91 per year. One possible explanation is a change in filing practice rather than in research output, but the data in this report does not distinguish the two.

Measurement is the single largest field with 415 families, 18.9% of the portfolio and 58% larger than the next field. At sector level the portfolio is unusually even: chemistry (808), mechanical engineering (770) and instruments (742) are within 9% of each other, and only electrical engineering (578) and other fields (139) fall clearly behind. The aggregate therefore has no single technological centre — the specialisation appears only when the institutions are looked at individually.

The individual portfolios are far more concentrated than the aggregate. Materials and metallurgy accounts for 48.7% of TU Bergakademie Freiberg's families, biotechnology for 38.9% of Universität Leipzig's, optics for 36.2% of FSU Jena's and measurement for 34.9% of TU Ilmenau's. TU Dresden is the exception: its largest field, measurement, reaches only 17.7% of its own portfolio — the profile of a filer whose volume spans the whole technical spectrum rather than one field.

Measurement is the connective tissue of the region. It appears in the top three of five of the nine institutions shown — TU Dresden, FSU Jena, TU Ilmenau, Universität Leipzig and HTW Dresden — and is the largest field for two of them. No other field recurs across as many institutions.

771 of 2,197 families (35.1%) carry at least one external co-applicant, and Fraunhofer alone accounts for 251 of them — 32.6% of all externally co-filed families and 11.4% of the entire portfolio. The next-largest single partner record reaches 22. No other relationship in this dataset comes close.

The institutions collaborate outward far more than with each other. Only 39 families (1.8% of the portfolio) join two or more of the seventeen institutions, against 771 with an external partner. The densest internal link is between FSU Jena and Universitätsklinikum Jena (10 families), which is consistent with their being one academic site under two legal entities.

Fraunhofer is by a wide margin the largest non-university research applicant in both states — 481 EP families with a Saxon inventor and 273 with a Thuringian one, roughly 4.6 times the Leibniz figure in Saxony. It is also the dominant co-applicant of the universities themselves (251 joint families), which places the same organisation at both ends of the regional research relationship.

The two states have different non-university profiles. Helmholtz is a Saxon phenomenon in this data — 77 families against 10 in Thuringia, consistent with the Dresden-Rossendorf site — while DLR is the only one of the five with more Thuringian than Saxon families (9 against 6). Leibniz is the most evenly distributed of the large organisations (104 / 59).

Four in five families take the German national route — 1,763 of 2,197 have a DPMA filing — which is the classic German university priority-filing pattern. Beyond that, 755 families carry a PCT application filed by an institution itself, and 605 reach the EPO. Counting all members of a family, not only those the institutions filed, 876 families (39.9%) contain a WO member.

The portfolio internationalised between the two complete windows. The DPMA share fell 5.5 percentage points while EP gained 7.4, WO 4.8, the US 3.5 and China 3.1 — and China doubled in absolute terms, from 34 to 55 families, in a window whose total family count fell from 1,117 to 906. The pattern is consistent across every non-German office in the table; no office moved in the opposite direction.

The European route grants more often than the German one — 59.6% at the EPO against 47.1% at the DPMA for the same 2013–2020 cohort. The United States is higher still at 67.5%. One consistent reading is selection: a family only reaches the EPO or the USPTO after the applicant has decided it is worth the additional cost, whereas the German filing is often the first and cheapest step. The data shows the difference but cannot establish the cause.

Across the twelve institutions with a reportable cohort the grant rate ranges from 29.3% to 76.5%, against an overall 50.2%. The two life-science heavy portfolios sit at the bottom — Universität Leipzig 29.3% and Universitätsklinikum Jena 38.0% — while TU Bergakademie Freiberg (57.8%), HTW Dresden (59.1%) and EAH Jena (57.7%) sit at the top. WH Zwickau's 76.5% rests on 17 applications and should be read as an indication only.

Eight of the twelve most prolific inventors work at TU Dresden, which matches its share of the portfolio. What differs is weight: 49 families is 4.7% of TU Dresden's 1,032, while Christos Aneziris's 28 families are 11.9% of TU Bergakademie Freiberg's 236 and Jens Limpert's 23 are 10.3% of FSU Jena's 224. In the smaller portfolios a single research group carries a visible share of everything the institution files.

1,228 of 2,197 families (55.9%) have been cited at least once, together receiving 9,225 citations. The distribution is steeply skewed: the 101 families with 20 or more citations are 4.6% of the portfolio, while 969 families have never been cited. An average of 7.5 citations per cited family therefore describes almost none of the individual families.

Citation intensity does not follow portfolio size. Universität Leipzig has 95 families but averages 14.8 citations per cited family, nearly twice TU Dresden's 7.6 across 1,032. FSU Jena has both the highest cited share among the large filers (64.7%) and an above-average intensity (8.6). At the other end, HS Mittweida's 71 families produced 45 citations in total — a portfolio filed almost entirely at the DPMA, where fewer citing documents exist to be counted.

Dresden alone accounts for 43.0% of Saxony's EP families (3,086 of 7,171) and Jena for 31.9% of Thuringia's (1,346 of 4,226). Both are the seats of the largest filers in this report. Beyond the four large cities the distribution is remarkably flat: eight further districts lie between 374 and 614 families, with no second cluster.

Academic applicants are a small share of regional EP output. Universities appear on 6.9% of Saxony's and 5.2% of Thuringia's EP families, and government or non-profit research institutes on a further 9.4% and 8.5%. Companies carry roughly seven in ten in both states. The academic contribution is best read as a research input into a company-dominated regional portfolio rather than as its main body.

Data currency. All figures in this report were computed on 15 August 2026 against EPO PATSTAT Global, Spring 2026 (patstat-mtc.patstat_2026a), with classification titles from IPC edition 20260101 and CPC edition 202608. Patent counts shift with every PATSTAT release; figures computed on a different edition are not directly comparable with these.