3D concrete printing dataset of linked in-line sensory data and interlayer bond strength

The dataset contains in-line sensory data and their corresponding off-line destructive test results for interlayer bond strength and other mechanical properties, collected during extrusion-based 3D concrete printing (3DCP). It includes measurements of 1448 printed specimens locations across 39 layers each, of which 1028 were subsequently saw-cut and mechanically tested. Detailed descriptions of the sensor systems are provided in [1, 2], while information on test procedures, data fusion, and the experimental program can be found in [3]. Using a multi-sensor process monitoring setup, data is acquired on robot motion, layer and object geometry [1], material preparation and transport [2], surface dehydration [2], and atmospheric conditions [2]. For each sensor, the raw measurements were transformed into a concise set of physically interpretable features using domain knowledge. After printing, specimens were extracted from the printed elements and tested in flexure to quantify interlayer bond strength, while compressive strength and flexural modulus were also derived from the same specimens [3]. All features were fused into cohesive digital representations of the printed objects, and the measured mechanical properties were co-registered with the corresponding in-line measurements [3], forming the basis of this dataset. In addition to the features and mechanical properties, the dataset also includes several specimen classifications, following [3]. Collectively, this data structure supports the development of data-driven virtual sensors for in-line prediction of interlayer bond strength. The tabular dataset and its structure are described in detail in [3] (Section 5), including an explanation of the row indexing scheme and a comprehensive overview and visualizations of all included features. References J. Versteege, R.J.M. Wolfs, T.A.M. Salet, Data-driven additive manufacturing with concrete: Enhancing in-line sensory data with domain knowledge, Part I: Geometry, Automation in Construction, (ISSN: 0926-5805), 172 (2025) 106020, 10.1016/j.autcon.2025.106020 J. Versteege, R.J.M. Wolfs, T.A.M. Salet, Data-driven additive manufacturing with concrete: Enhancing in-line sensory data with domain knowledge, Part II: Moisture and heat, Automation in Construction, (ISSN: 0926-5805), 177 (2025) 106327, 10.1016/j.autcon.2025.106327 J. Versteege, R.J.M. Wolfs, T.A.M. Salet, Sensory data mapping and database development for interlayer bond strength: Towards digital shadows in 3D concrete printing, Additive Manufacturing, (2026), 10.1016/j.addma.2026.105105