Sustainable construction work is subjected to diverse requirements. Material, finance and manpower resources should be employed sparingly; a construction site ought to generate little disturbance for others; durable, long-lasting structures need to be created [1]. Monetary criteria are not sufficient of themselves. A holistic objective would rather be to avoid the wastage that can characterise all the criteria listed here. Current construction work in Europe on new bridges and high-rise buildings shows how important time factors have become. Slow work at a construction site is not in harmony with sensitively networked goods and logistic flows and can become particularly wasteful when traffic jams, emissions and lost human working hours are considered. The objective of Priority Programme (PP) 2187 “Adaptive modular building design with flow production methods” is to develop procedures for prefabrication in series that facilitate the shortest construction time possible [2]. To this end, structures are divided into similar modules, prefabricated in a factory and then just joined up together at a construction site. It is important to track the modules with sensors and assure their quality without fail so that no component is lacking, no post-processing is needed and that they are erected according to the design of their digital twin and precisely reflect this digital base model.

Peter Mark and Patrick Forman, Ruhr University Bochum, Germany

The German Research Foundation (GRF) began sponsoring this PP in 2020. The programme is scheduled over six years and brings together researchers from the disciplines of structural engineering, production technology, construction informatics and mathematics. Cross-discipline thinking and developing are required. Eight university sites are involved: the University of Stuttgart, the Universities of Applied Science from Munich, Dresden, Berlin and Chemnitz, the Karlsruhe Institute of Technology, the Leibniz University of Hanover and the Ruhr University Bochum.

The programme content deals with three main fields of research:
- Design and construction in terms of modularization
- Manufacturing strategies and production concepts for scalable structural modules
- Continuous digital models for designing, manufacturing and assembling processes

We are dealing with modularization methods and module concepts, e.g. dismantling concepts, joining principles or connections with tolerance compensation. The main focus is on developing production systems, quality assurance and production management in the area of manufacturing and production. The objective with the digital models is to be able to derive consistent interaction modelling and thus continuous linking of relevant processes including their implementation in a utilisation phase.

The illustration shows the different research subjects of the 13 sub-projects assigned to three fields of research (coloured circles). They have been almost uniformly distributed amongst the fields and are, as a general rule, undertaken as tandem projects by two research institutes. An introduction will be made to the content of all these sub-projects in short articles in the following editions of CPI – Concrete Plant International.

Literature
[1] Ahrens, M. A.; Strauss, A.; Bergmeister, K.; Mark, P.; Stangenberg, F.: Lebensdauerorientierter Entwurf, Konstruktion, Nachrechnung, Betonkalender Bd. 1, Hrsg. Bergmeister, K., Fingerloos, F. & Wörner, J.-D., 2013, S.17-222. (ISBN 978-3-433-03000-4)
[2] Forman, P.; Mark, P.: DFG Schwerpunktprogramm 2187: Adaptive Modulbauweisen mit Fließfertigungsmethoden – Präzisionsschnellbau der Zukunft, BetonWerk International 85(6), 2019, S. 12-14.

Sub-project SPP 2187

Prof. AA Dipl.(Hons) Achim Menges, Dipl.-Eng., LEED AP Tobias Schwinn and Dipl.-Eng. David Stieler, University of Stuttgart, Institute for Computational Design and Construction (ICD), Germany

Agent-Based Methods for Fabrication-Oriented Design of Adaptive Modular Prefab Concrete Construction

Highly efficient, load-adapted structures feature a great deal of differentiation in their construction elements. According to the principle “Large scale individuality – Small scale similarity”, this variability will be able to be implemented using adaptive basic modules under mass production conditions. Computer-based design methods, which facilitate both adaptability as regards construction element geometry and continuous manufacturing-oriented designs, are necessary in order to open up the potential of adaptive modular building design with flow production methods in respect of rapidity and quality but with flexibility for building structures at the same time.

Investigations at the ICD Stuttgart are thus focussed on developing continuous digital design methods for modularized concrete structures. Multi-criteria aspects as regards segmentation, joints, manufacturing and assembly will be taken into account and be exchanged. Continuous digital models will be able to satisfy the demands of the design and planning process as well as form the basis for digital production. Yet, the focus will not just be on continuity but more especially on integration using systematic feedback from technical parameters in the design with planning phases that usually follow on chronologically. About 80% of the environmental and economic performance of a completed building is determined by decisions in design that will have been taken in the first 22% of the entire planning time [1]. For this reason, feedback in the planning process will not occur sporadically but rather iteratively and with high-frequency in rule-based digital “loops”. Apart from these criteria, particular attention will also be paid to interfaces concerning structural design, life cycle and lifetime cost analyses.

Agent-based modelling (ABM) is proposed in order to transpose the segmented character of design, planning and production into a continuous process that is bound together. This involves computer simulations, in which, for the most part, a number of individuals (agents) interact in an environment [2]. It allows complex interdependencies to be illustrated far beyond planning methods based on databanks or parameters. A great deal of previous work at the ICD Stuttgart in the area of integrative, computer-based design has already demonstrated the usefulness and advantages of ABM in comparison with manual or even parametric design methods [3].

The ABM method will be expanded in respect of flow manufacturing for precast concrete element structures by building on this methodical framework. A synthesis of the requirements with this agent-based modelling approach will be initially generated as a generally applicable planning methodology for digitally manufactured precast concrete elements, before the method is validated and evaluated in the context of a specific case study. Optimising agent behaviour by means of reinforcement learning will also be a substantial new contribution to this project. Machine learning will be employed with the ABM in order to better predict desirable system properties at a macro-level and be able to more rapidly develop clearly focussed rules for decision-making.
In future, this method will not just be relevant as an integrative planning approach but also as a management and control procedure within a cyber-physical production system.

Literature
[1]  Bogenstätter, U. (2000). Prediction and optimization of life-cycle costs in early design. Building Research & Information, 28(5–6), 376–386.
[2]  Metz, T. (2017). Agent-Based Modeling (ABM). In Sebastian Jäckle, editor, Neue Trends in den Sozialwissenschaften Innovative Techniken für qualitative und quantitative Forschung, S. 11-12.
[3] Groenewolt, A., Schwinn, T., Nguyen, L., Menges, A. (2018), An interactive agent-based framework for materialization- informed architectural design. Swarm Intelligence, Volume 11, Special Issue on Self-Organised Construction, Springer.

Sub-project SPP 2187
Albert Albers, Christoph Kempf, Robert Renz and Markus Spadinger, Karlsruhe Institute of Technology (KIT), IPEK – Institute of Product Engineering, Germany
Lothar Stempniewski and Agemar Manny, Karlsruhe Institute of Technology (KIT), Institute of Concrete Structures and Building Materials (IMB) – solid structures department, Germany

Clever modularization for scalable concrete construction through adapting methods of developing modular construction kits

Modular construction work facilitates product and process standardisation in manufacturing construction elements. This construction method additionally forms an important basis for exploiting economies of scale with differing building sizes and types. Utilising ultra-high-strength concretes with carbon/textile reinforcement also makes it possible to erect long-lasting, more lightweight buildings. Such concretes are a cost-effective, environmentally-friendly alternative to classic steel reinforced concretes. The objective of the joint research project is the transfer of methodologies from mechanical engineering to civil engineering for developing construction kits and platforms as well as investigations into the ultra-high-strength textile concrete components of construction kits. Individual construction elements will be assembled in situ using a zero tolerance dry joint – as is known from modern vehicle manufacturing. Standardised modules can be produced and assembled efficiently with savings on costs and time. The skeleton construction method shown in the illustration will be employed in carrying out segmentation. This skeleton method is a means of construction suited to construction kit architecture due to its great flexibility in terms of floor plan and elevation.

In this research project, the solid structures department at the Institute of Concrete Structures and Building Materials (IMB) at the KIT is focussing on developing a basic concept for erecting non-flexural frame structures from individual construction kit modules in defined product series. The prefabricated elements are to be prestressed to create an integral load-bearing structure with the aid of prestressing tendons. In this case, the questions to be addressed are: how structural requirements as regards force transmission at the zero tolerance joints can be assured, how the interfaces between tendon and construction element can be formed in a standardised way as well as how the textile reinforcement can be correctly positioned in the construction element and how these high-performance construction elements should be dimensioned.

The focus for the IPEK – Institute of Product Engineering at the KIT – is rather on the transfer of methods from construction kit and product series development to construction industry development methods in order to make their advantages in efficiency and effectiveness available to civil engineering. An evolutionary algorithm will define the structure and building blocks for the construction kit in order to determine optimum model variations whilst maintaining the objective of minimum internal diversity alongside great external diversity. Here the questions to be asked are: what structure must the building kit exhibit, which modules are necessary and what dimensions are possible. Fulfilling the sub-goals in this proposed joint research project will make possible short construction times almost independent of the weather accompanied by a substantial reduction in material. All this will generate appreciable savings on construction costs. Interdisciplinary cooperation between the two institutes will promote an exchange of knowledge and transfer between the domains of civil and mechanical engineering.