Lars Stehn

Applicability of lean principles and practices in industrialized housing production

The applicability of lean principles and practices to industrialized housing in Sweden are examined, taking the production culture into consideration. The factory production of industrialized housing shows apparent similarities to manufacturing, but areas related to fully integrated lean production practices, such as error proofing and standardized work floor and equipment maintenance, are scarce. Hence, applicability of lean principles and practices to industrialized housing production is clearly influenced by a production culture that has similarities to a traditional construction culture. Setting up industrialized housing production thus requires careful implementation of lean principles if workers from traditional building are moved into factories, and managers still adhere to the prevailing site‐based production mentality. However, the influence of the traditional construction project culture is not solely a constraint; flexible teams that take their own responsibility are also important in a lean culture. Hence, retaining parts of the existing construction mentality, context and way of working is also central when discussing lean applicability in industrialized housing.

Integrated design and production of multi-storey timber frame houses – production effects caused by customer-oriented design

Abstract This paper presents the results from an ongoing R&D project aiming for an industrialised development of a multi-storey timber frame house system. The development was conducted systematically using methods from concurrent engineering (CE) focusing on the customer satisfaction and production–design problems/possibilities. A Swedish SME building company was investigated from the viewpoint of customer orientation, production and design of multi-storey timber frame houses. The company uses industrial fabrication of volumes. The aim was to analyse how the production can benefit from an integration of a customer-oriented design and production. First, a model based on the QFD method is proposed on how requirements from customers, i.e., tenants, can be taken into account in the design process and how CE can be adapted to a SME design process. Second, a preliminary model is proposed considering disturbances and relative cost effects on the production due to changes in the design solutions. The presented model aims to predict the total cost for a customer affected design on the used building system to forecast the costs for the main contractor and building owner for a similar integrated design in the future. All observations are based on case studies of the design process and the industrialised production of a three-storey timber frame house with three different floor designs.

Construction clients’ ability to manage uncertainty and equivocality

While risk and uncertainty management have received much attention within construction management research, management of information interpretation remains unexplored. Situations that are generally overlooked are those where increased amounts and flow of information are not the solution to the human problem of managing multiple meanings of information and conflicting interpretations, i.e. ‘equivocality’. The aim is to identify and differentiate between construction clients’ uncertainty and equivocality about industrialized construction in Sweden, and, in the light of those findings, to evaluate clients’ current information processing practice on investment decisions in new-build in order to assess and discuss clients’ ability to manage uncertainty and equivocality. Based on information processing theory, analysis of aggregated data from three previous studies shows that there is a need to manage both uncertainty and equivocality. At the same time, clients’ ability to do this is found to be limited. Consequently, when industrialized construction moves clients beyond their current frame of reference, clients’ information processing practice does not support decision making. It is also proposed that differentiating between uncertainty and equivocality will enable a more profound understanding of the sequential order for information processing, i.e. that one must define questions (reduce equivocality) before one can find answers to the questions (reduce uncertainty).

A lean-agile model of homebuilders' production systems

A lean‐agile model of homebuilders’ production systems is proposed in this research. Value stream mapping is utilized to clarify the proposed lean‐agile model, in which pull and Kanban are used to ensure smooth production upstream of the de‐coupling point, while fluctuating market demands in terms of variety of homes and variability of volumes are managed by the agile process downstream of the de‐coupling point. The de‐coupling point is used to provide components to downstream agile process and shield upstream smooth production from market fluctuations. A simulation model is developed to evaluate and validate the effectiveness of the proposed lean‐agile model. Simulation experiments show that the lean‐agile model prevents the accumulation of high inventory levels and thus provides better customization opportunities for clients compared to even‐flow‐construction. It also provides a more stable process with shorter cycle times compared to sales‐driven production. The proposed lean‐agile model offers new possibilities for homebuilders to manage the balance between meeting fluctuating market demands and stabilizing the production system.

Benefits and disadvantages of ERP in industrialised timber frame housing in Sweden

Industrialised timber frame housing is successful in the Swedish market for one‐family housing. In the manufacturing industry, methods and software systems such as enterprise resource planning (ERP) are widespread, demonstrating significant benefits. A survey of the majority of industrialised timber frame housing companies in Sweden demonstrates low ERP use with a low degree of strategic importance, but with operational and managerial benefits. The ERP approach has potential for industrialised housing and its use is favoured by an increased maturity in IT.

The impact of clients' decision-making on their adoption of industrialized building

Previous research has shown that client organizations’ decision-making about new-build creates individual-level and organizational-level barriers to the adoption of Swedish industrialized building (IB). However, it has been proposed that clients may overcome barriers on both levels by allowing multiple meanings and conflicting interpretations to surface and interact with their decision-making. The aim is to test this proposition. Based on the theoretical fields of decision-making and organizational information processing, a framework for analysis has been developed. In the framework, three decision-making approaches are operationalized: rational, judgments and managing multiple meanings. Data were collected using in-depth interviews with key decision-makers from four Swedish professional clients differing on when and if they adopted IB: one early-adopter, two late-adopters and one non-adopter. The empirical findings support the proposition and show a relationship between how clients manage multiple meanings in their decision-making and their adoption of IB. The research adds to the understanding of how clients may overcome barriers to the adoption of IB on both individual and organizational levels. Moreover, it increases understanding about how clients might better cope with radical changes and innovations.

A design structural matrix approach displaying structural and assembly requirements in construction: a timber case study

A fundamental demand of construction design is human safety from structural failure. As a consequence, buildings generally tend to be structurally optimized with cost as the main target parameter. However, a cost-suboptimized structural design often leads to poor constructability decisions with subsequent waste. This paper presents initial research in the development of a design structural matrix (DSM) method able to identify constructability obstacles between structural design and assembly and thus eliminate waste. Empirical data based on a case study of long-span timber structures is used in the development and analysis of the method. The DSM was found to be a holistic tool for systematic consideration of structural design and constructability requirements by providing a standardized system view, a detailed element view, and physical and functional interactions among elements and modules. The DSM was also shown to aid in detailed design and production management through the use of simple matrix tools.

Plug shear failure in nailed timber connections: Load distribution and failure initiation

Brittle failures in mechanical timber joints should be avoided, because this often results in low capacity and brittle failure of the structure. Nailed joints experience three ultimate failure modes: embedding, splitting or plug stear failure. To avoid plug shear failure, short and wide joints are preferred, limiting the number of fasteners in line with the load and grain direction. Plug shear failure was examined in short-term experiments on nailed steel-to-timber joints in glulam loaded in tension parallel to the grain with five different joint geometries. The aim of the study was to examine if the fastener group layout can be adjusted to avoid plug shear failure and to gain an insight into the causes of failure initiation. Using spring models, it is shown that the load distribution creates pronounced stresses at the last nail in the joint, which probably initiates the plug shear failure. Test results are compared with prediction models found in the literature. It was found that fasteners placed in groups can be a successful way of reducing the risk of plug shear failure. The failure is probably initiated at the nail farthest from the free end, where tensile stresses perpendicular to grain occur.ZusammenfassungSprödbruch in mechanischen Holzverbindungen sollte vermieden werden, da dies vielfach zu einer geringen Tragfähigkeit und Sprödbruch in der gesamten Konstruktion führen kann. In genagelten Holzverbindungen können insgesamt drei Brucharten auftreten: Bruch durch Überschreitung der Lochleibungsfestigkeit, Aufreissen, oder Ausreissen des gesamten genagelten Bereichs. Um solche Scherbrüche zu vermeiden, sind kurze und breite Verbindungen zu bevorzugen, bei denen die Anzahl der Verbindungsmittel in Last- und Faserrichtung begrenzt wird. Das Schubversagen wurde in Kurzzeitversuchen an Holz-Stahl-Nagelverbindungen von Brettschichtholz bei Zugbeanspruchung parallel zum Faserverlauf mit fünf verschiedenen Verbindungsgeometrien geprüft. Ziel der Untersuchung war, festzustellen, ob die Nagelanordnung so gewählt werden kann, dass Schubversagen vermieden wird, und um Einblicke in die Ursache von Bruchauslösung zu gewinnen. Mit einem Federmodell kann gezeigt werden, dass die Lastverteilung erhebliche Spannungen am letzten Nagel in der Verbindung verursacht, wodurch aller Wahrscheinlichkeit nach das Schubversagen entsteht. Versuchsergebnisse wurden mit theoretischen Modellen aus der Literatur verglichen. Es stellte sich heraus, dass gruppenweise angeordnete Verbindungsmittel das Risiko eines Schubversagens deutlich vermindern können. Wahrscheinlich beginnt der Bruch an dem vom freien Ende am weitesten entfernten Nagel, wo Zugspannungen quer zur Faser auftreten.

An indicator for superfluous iteration in offsite building design processes

Enhancement of iteration management in the design phase is important for successful offsite building projects. Design iteration has two aspects. Although iteration is necessary to deal with design requirements when solving complex problems (i.e. increasing quality through iteration), it has also been identified in numerous studies to be one of the main causes of design errors and time and cost overruns (i.e. superfluous iteration), as it increases scheduling and design complexity. Current building project management tools do not provide a means to control the reduction of superfluous iteration. One problem is that existing research has difficulty precisely relating the effects of specific management actions to superfluous iteration. The idea of this study is to develop an indicator, the sequence deviation quotient (SDQ), which reflects the amount of superfluous design iteration in a project. It can be thought of as a tool supporting project managers to make systematic and continuous (from project to project) design process improvement. A premise is that the impact of varying project conditions on the process structure of design processes, i.e. the precedence relationships between the design activities, is only small. In this paper, we provide a definition of superfluous iteration. We tested the feasibility of the SDQ by subjecting it to project variation and input perturbation by means of a Monte Carlo simulation. The simulations are based on the data from a real offsite design building process, the designing of a 1100 m2 residential building in Sweden.

3D Computer Visualization in Timber Construction—Some Important Parameters

Abstract Timber building structures are seldom considered because of a lack of timber construction knowledge among many clients and professionals. 3D computer visualization (3D VIZ) is a possible way of communicating the aesthetics of a visible timber structure and visualizing complex timber connections. This paper investigates the potential for 3D VIZ to communicate visible timber structures during the different phases of the construction process. Furthermore, important parameters (controllable in 3D VIZ) for communicating a timber structure are identified. Through an Internet-based Delphi study, the usefulness of 3D VIZ for timber structures has been evaluated by professionals representing different competencies in the construction process. The results show that structural complexity, intended beholder (professional or non-professional) and current construction phase influence the benefit of 3D VIZ. The level of detail in the visualization is heavily dependent on whether the communication is internal (between professionals) or external (between professionals and clients/users). The impact of parameters needed, such as textures, surface structures and realistic lighting, have been described.