|
To get the best out of this structured learning exercise, you should start at the beginning.
- Learning objectives
After working through this section, you should be able to:
- understand the driving forces which motivate some organisations towards developing construction robots and automation
- recognise the existence of inhibitors to the development and deployment of construction robots and automation and the means for overcoming them
- understand the issues in relation to R&D and the approach to be taken when embarking upon a programme involving construction robots and automation
- appreciate the main differences between robots and construction plant and equipment in terms of their basic engineering
- identify the different generic types of construction robot
- appreciate the development paths and potential for convergence of these types of construction robot
- recognise the achievements to date in terms of real developments and the regions of the world where these have occurred
- appreciate how the future might unfold in terms of the form of construction robots and automation
- recognise the impact that newer forms of construction automation are likely to have on the construction industry.
At the end of this self-study exercise, you will be able to test your understanding of the issues in relation to construction automation and robotics by attempting a selection of questions. Personal feedback is give to you in the form of a separate answers' page.
- Introduction
In this section, we introduce the concept of the construction robot and the various approaches taken to date towards automating on-site production processes. It is important to recognise that the word robot, in the context of construction plant and equipment, is used extremely loosely. We might be more accurate if we said robot-like, as there are very few devices which could be classed strictly as robots. Automation is a word that is similarly misused. Nonetheless, these terms have become so widely accepted that it would serve no useful purpose to debate their meaning here.
We begin this section by discussing the origins of robots in the construction industry and their impact upon the overall process. Subsequent parts of this section deal with the motivations of organisations in regard to the development of robots and the inhibitors to their deployment on construction sites. You will be reminded of how progress around the world varies from region to region and how the Japanese - the leading proponents of this technology - are embarking upon ambitious plans to automate the construction site. This section will outline the agenda of current R&D and will promote a vision for the future which you will be asked to consider in the context of your own organisation's work, should that be an appropriate situation to consider. Later sections to be published on this web site will deal in more detail with technical aspects. Our primary aim in this section is to provide a broad view, but a developed understanding nonetheless, of the core issues and directions in which construction robots and automation are moving.
- Review of the origins of robotics for construction
Many organisations and their countries would lay claim to having developed the first construction robot. The prize for having been the first could be awarded to the British, Americans, Germans, Japanese or Russians. No one quite knows who produced the first nor does it really matter. What we need to understand is that a number of countries have attempted something and not, necessarily, for the first time.
There have been many attempts at automating parts of the process, dating back to the time of Brunel and his contemporaries. In his building of the Crystal Palace for the Great Exhibition, Paxton engaged in so many innovative production practices that only today we seem to have caught up. Fabrication, assembly and erection processes that used machines instead of men were early forms of automation. In the late 1970s, masonry robots, capable of laying regular bricks and blocks, were under development in at least two construction companies in the UK. So, there is a long history to the broader setting of this subject. Why these UK initiatives in particular, and others of that era, failed to bear fruit will become apparent later in this section. More important at this point is that we recognise the willingness of the construction industry to innovate, but also its ignorance of the technological and other barriers to be overcome. This spirit of inventiveness is not echoed today in the latest R&D programmes, most of which are painstakingly thorough developments relying on substantial long term funding. In this respect at least, the age of the workshop inventor may have gone forever.
Activity 1
Cast your mind back to when you first saw either a prototype or working construction robot. Try to remember what the task was and your perception of its effectiveness. Key in your response; you can print (or save) this section for later reference.
Task involved:
Perception of effectiveness:
- Driving forces for construction robots and automation
There can be many reasons why a company should seek to develop robots and automation. The primary motivation ought to be obvious: the machine is meant to replace the man. Derived from the Czech word 'robota' for forced work or slavery, a robot could be used for work which is hazardous for a worker to perform. Generally, the development of robots has been justified by organisations on the grounds of productivity, safety and quality. For the construction industry, a primary motivation has been the prospect of gaining competitive advantage through lower input costs. Thus, the driving force has been mostly economic. Other factors are, however, evident especially concerns over labour shortages.
In Japan, labour-substitution, safety and quality issues are the main motivations. In a country where, for all practical purposes, full employment has been the norm, attracting people to work on construction sites is not easy. Moreover, it would be fair to say that Japanese construction companies prefer not to bring in foreign workers to their sites. This has to be contrasted to the UK where the use of casual labour at the lowest cost has been the main defence and weapon in dealing with stiff competition. Even so, the motivations are often more complex and interact in quite different ways from construction company to construction company and from country to country.
We might consider the main driving forces as those given below.
- Productivity and labour
Higher output at lower unit cost
Improved competitiveness (especially internationally)
Shortage of skilled workers/craftsmen
Growing number of older workers
Unattractiveness of construction as a career for young people
Lesser dependence upon casual workers
- Safety and quality
Avoidance of work in high, dirty, unpleasant or dangerous places
Relief for, and protection of, machine operators
Safer operation of machines
Better execution of the work itself (process improvement)
Greater consistency in the outcome of the work (product improvement)
Higher level of control over production processes generally
The main driving forces for construction robots and automation
Activity 2
If your company (or one you might know) were to embark upon an R&D programme leading to the development of construction robots and automation, on its own or in collaboration with others, what would be its main motivations? Try to think in terms of both tangible (i.e. quantifiable) benefits and intangible (i.e. qualitative) benefits.
Box 1: Tangible benefits; Box 2: Intangible benefits:
- Inhibitors to progress
The low level of use of robots and automation and, in most cases, the absence of anything resembling them on construction sites ought to indicate the presence of barriers or inhibitors. Given the established use of robots and automation in many other industries - automotive and electronics component manufacture are obvious examples - there have to be reasons for the low level of application of this technology within construction. There are, indeed, many reasons and some have nothing to do with the technology itself.
Activity 3
Try to imagine the attributes that a robot might possess in order for it to be able to adapt to the range of tasks found on a construction site. You might wish to differentiate between physical and cognitive attributes (intelligence or reasoning).
Attributes or characteristics of a construction robot:
Box 1: Physical; Box 2: Cognitive
There is broad agreement amongst researchers and practitioners on the main inhibitors to the deployment of robots and automation on construction sites. These can be grouped variously under the headings of economics and market, structure and organisation and technology.
- Economics and market
It should be self-evident that the cost of developing technology of the kind discussed here represents a substantial investment and risk for anyone. If and when this is forthcoming and the new device is ready to take its place on site, how will it justify its cost and generate a return for the investor? For anyone contemplating R&D on the scale required for the successful introduction of construction robots and automation, the financial case has to be well proven with the risks properly considered. The latter is particularly important in the context of technological innovation. More likely than not for construction, technology can be borrowed from other industries where the high cost of initial development has already been borne. Technology transfer is something that has to be evaluated by those contemplating product development. Making one's solution adaptable to the requirements of other markets can increase sales many fold. There will, however, be few instances where a robot developed for construction could have benefits which would spin-off into other industries unless, of course, that device were to have solved problems yet to be resolved in other industrial applications.
- Case Study: The UK's Advanced Robotics Initiative
In the late 1980s, a group of UK construction companies and other industrial firms collaborated in, first, a feasibility study and then in a project definition study for an advanced robot. This programme of R&D was supported by the Department of Trade and Industry under its advanced robotics initiative which had identified civil engineering as one of its key applications' areas. An initial (scoping) study had been conducted by CIRIA (Construction Industry Research and Information Association) and this confirmed the industry's interest in pursuing a feasibility study for a device that could inspect, test, repair and maintain large or tall structures. Typical of the target applications were structures that were generally inaccessible or difficult for workers to reach. Power station cooling towers, chimneys, soffits of bridges and suspended carriageways fell neatly into this category. Moreover, the cost and time in manual inspection alone was considered a sufficient motivation to justify further examination. For the device to be supported financially by the DTI, it would have to conform to the definition of advanced. In other words, the device should have a large degree of autonomy making it possible to navigate and sense across the face of the target structure, probing as necessary to reveal where problems might be found.
The scoping study had been conducted over a few months and was understood to have cost a few tens of thousands of pounds sterling. The feasibility study, in reconfirming and then adopting the recommendation of the CIRIA report, embarked upon an evaluation of the various issues that a development project would have to address. This was not, therefore, going to be development in itself, but it would require a significant theoretical framework to be designed and tested. Additionally, the economics and market for the device had to be assessed, a task made difficult by having to forecast demand without knowing what could be supplied. The feasibility study was supported by DTI to its highest level of funding - 75% of real costs - over a period of about 12 months. The final cost of this stage was several times greater than the scoping study. Thus, moving from largely desk-based work had increased the cost by an order of magnitude.
As the work proceeded, it became obvious that there were many problems that had to be solved. Not only was the market for such a device difficult to forecast, but the technological complexity associated with it proved progressively more challenging. Even so, the study did conclude its work and was able to show that an inspection and testing robot for civil engineering structures was, indeed, feasible. There was, however, a difficulty in balancing the requirement for an advanced (i.e. autonomous) device and the need to use proven technology. There was also the very real prospect that the device might not be allowed to roam freely across a structure and that it might have to be tethered. The feasibility study led eventually to a project definition study which would in effect build the prototype device. For various reasons and one imagines cost played a large part in this, the prototype has yet to be built.
The feasibility study had concluded that the cost of the first production run might be in the region of a few million pounds, depending on functionality. That said, costs would not necessarily fall to the level that many people had hope for in the early days. In the event, too many problems - economic and technological - conspired to deny the UK of its own robots to sell to the rest of the world. There are many lessons which could be learned from this programme. Perhaps the more obvious are that the project was overly ambitious and that it was being promoted at a time when the construction industry was falling deeply into recession. Whatever the full extent of the reasons, there can be no doubting the enormity of the effort involved in delivering construction robots to the marketplace.
- Structure and organisation
The fragmented state of the construction industry is blamed for many ailments including one of the lowest rates of investment in R&D across all industries. We know that, conversely, this state permits a high degree of flexibility in its methods to meet the needs of its clients and customers who are many and diverse. We know equally that where the responsibility and control is split between different parties and where no one organisation can take overall charge of the process, innovation can easily be stifled. Moreover, continuity in production - a feature which other industries must have if they are to produce competitively - is largely absent from construction. These would appear to be real reasons for most countries to defer on R&D expenditure. This does not, however, explain why the Japanese have continued to pump large sums of money into the development of construction robots and automation.
Reasons here are many fold and have much to do with cross-ownership of companies which is so prevalent in Japan. We might like to think that most of the reasons have to do with money. That misses one obvious factor: the large Japanese construction companies exemplify the principle of single point of responsibility . By exercising control over much of the process and its many and different contributors, they are able to undertake R&D at lower risk and with a higher expectation that the results will have worthwhile application on their construction sites. The Japanese have also understood that to invest in one's suppliers (taken in the broadest sense) is to invest in one's own success. Additionally, the construction companies are more inclined to collaborate outside their own specialisation and to fund and manage R&D jointly with others.
Shortage of funding and support from government, which is often blamed by industry in the west, is not an issue in Japan. True, the Japanese benefit from taxation concessions, but it takes more than that to innovate. Believing that government subsidy will remove the greatest inhibitor and allow real investment in R&D to begin misses the point.
We need to understand these issues if we are to see later how current and future developments in wholesale automation of the construction process is possible in Japan. In doing so, we might also see how it can be possible in the UK and the rest of Europe. Below are factors which show how construction is different from other industries, in this case manufacturing, and why there are still barriers to the use of robots and automation on construction sites. If there were few distinctions between factories and construction sites, for instance, we would expect to see higher levels of investment in plant and equipment, including robots. This is patently not the case.
| Factor |
Manufacturing |
Construction |
|
| Location |
Work usually performed at one permanent location |
Work dispersed across many temporary locations |
| Product life |
Typically short to medium term |
Long service life which is greater than design life |
| Standardisation |
Highly standardised products or few variants tolerated |
Little standardisation with most buildings of unique design |
| Complexity |
Small reliance on manual skills, extensive use of automation |
Large number of tasks requiring extensive manual skills |
| Work area |
Tasks generally performed at static workstations |
Tasks performed over a large area with workers mobile |
| Ergonomics |
Workplace well adjusted to human needs |
Rugged, harsh work environment where application of ergonomics difficult |
| Work-force |
Largely stable work-force - low churn rate |
High turn-over of workers |
| Responsibility |
Unified design, production and marketing team |
Divided responsibility between customer, designers, constructors and specialists |
- Technology
Not all problems are due to a combination of economic and organisational factors. Technology is a more significant inhibitor to realising the goal of the automated construction site. Development of robots, where components are taken from other applications with little or no adaptation, can help bridge the gap between theory and practice. There are, however, some fundamental differences between the machines used in a factory and those used on a construction site. It is important these differences are appreciated in terms of their basic engineering. For the purpose of comparison, we shall use an industrial manipulator of the kind used to pick-and-place, weld or apply a finish and a multi-purpose machine such as the all-terrain, wheeled excavator/digger. The table below summarises the main differences.
| Characteristic |
Robot (e.g. industrial manipulator) |
Construction equipment (e.g.excavator/digger) |
|
| Physical strength |
Stiff, not rugged |
Flexible, rugged |
| Payload:weight |
High (1:20 or more) |
Low (1:5 or less) |
| Positional accuracy |
High (within microns) |
Low ( 100 mm) |
| Force feedback |
Unlikely |
Close coupling of operative and machine |
| Mobility |
Restricted |
Highly versatile |
- Design practices and processes
There have been few fundamental changes in construction technology over the centuries. It is well understood, with knowledge transferring easily from one generation to the next. The inherently crafts-based approach is, however, giving way to a process that is increasingly influenced by technology. Yet, most designs remain reflective of a production process that assumes the use of labour. For developers of robots and automation systems, this might appear to present an almost idea target and in certain respects this is so. Evidence is to be found in several examples of dedicated or task-specific devices that substitute effectively for labour. Stable technologies of the kind that are core to the construction process - concrete frames, masonry walls and wet-applied finishes - have long been the focus for developers of robots. As we shall see later, these technologies continue to be the focus for some.
- Targets for automation
In Japan, robots for power-floating concrete appeared around 14 years ago. The relative stability of the underlying technology - it has been around for decades and is likely to continue for some time - means that it is possible to produce prototypes under conditions of relative certainty. Shimizu, Taisei, Obayashi, Kajima and Takenaka (the so-called Big Five) have each developed a concrete power-floating robot, together with other devices to automate the concreting process. Positioning of steelwork has also been automated to some degree, as has spray-applied finishing. For these activities, robot solutions have been devised and implemented successfully.
This is not the case for other areas of building construction, amongst which engineering services and fitting out are the more prominent. Here the production process is complex and cannot be so easily rationalised. Even so, there are solutions, which from an automation perspective, are entirely valid. We are referring here to greater off-site prefabrication and the use of modular, standardised construction products.
In the case of engineering services - these can typically account for one half of the capital and three quarters of the running costs of a building - the complexity of the work invariably means that the scope for using robots on site is limited to lifting and positioning operations. On the face of it, this would appear to be a limited use of the technology. We should, however, see this differently. When safety and quality issues are considered, as they must, the welfare of the operative comes into full view. Minimising or eliminating physically demanding activities is a legitimate goal. Humans are adept at intricate tasks involving a high degree of dexterity, but weak at substituting for lifting machines. By using robots and automation systems to take the strain out of a potentially debilitating activity and relying on human skill to align and effect connections, a sensible and workable compromise is possible. In other words, we should concentrate our efforts on developing robots and automation systems to alleviate the strain and pain of assembly and moving components into position.
Many of the accidents on construction sites arise from some failure in materials' handling. Using humans to effect the fixings or connections and to do so from a safe and physically undemanding position has to make more sense than struggling to hold, position and connect the component all at the same time. Recognition of this is to be found in the heightened interest by Japanese companies in materials' handling and, with it, a re-focus of attention on organisation and management of on-site production processes. As alluded to earlier, dynamic workplaces, such as construction sites, create problems in logistics as well as in production. By making the site more like the factory, it is possible to solve several problems at once. As we shall see later, the factory concept is one that is fast coming to construction, especially in Japan. There is, however, more to this than better planning and control: issues like weather exclusion and design for automation have a key part to play too. Establishing a factory at one location for a fixed period, in order to create an environment that is more conducive to the use of robots and automation systems, is both logical and practicable. This is precisely what the Japanese are now doing through the agency of the leading construction companies.
- Need for rationalisation
Even without the advent of the construction robot, rationalisation of the construction process is needed. Most of the attempts so far in developing robot solutions have been largely task-specific. Furthermore, they have tended to replicate (as well as automate) manual processes which have largely been considered in isolation. In this regard, it is difficult to see how wholesale change of the kind that is needed in construction will occur from this attempt at innovation. The incremental approach to improvement which is so typical of the Japanese is unlikely to create the same transformation in construction as we have seen in other industries. Perhaps the underlying processes are so inefficient that the practice of kaizen will not deliver the measure of improvements necessary. Implicit in any study seeking to re-engineer on-site production process would be a re-examination of design, since it is that which drives production. Under procurement methods that separate design from production, the chances of any real improvement would seem to be limited. However, where procurement involves the practice of the single point of responsibility principle, there is a much better chance that rationalisation can occur. That condition would enable processes to be redesigned and technology, including IT, to be explored more purposefully.
Critics of this approach might argue that it weakens the architects' influence over the quality of design in the buildings so produced. The result, they further argue, will be buildings of unparalleled mediocrity. Yet, by bringing design, production and other specialisms closer together, a genuine form of integration can occur. Additionally, IT will enable the processes to be much more effective in delivering what the customer wants. Unencumbered by the limitations of separated responsibilities, single point of responsibility can invest in ways to automate the production process. The prospect of a harder technological edge to construction could be very real.
- Developments in construction robots
The development paths for construction robots and automation have occurred naturally and owe little to the intervention of governments, funding agencies or other external influences. Construction robots and automation fall into three categories: enhancements to existing construction plant and equipment; task-specific, dedicated robots; and intelligent (or cognitive) machines of which there are very few.
- Enhancements to existing construction plant and equipment
The development path has been largely overshadowed by the development of constructionrobots - see below. In many respects, it is the more important of the two, since it meshes well with the conditions that prevail on a construction site - see earlier sections on the differences between factories and sites and the engineering of industrial manipulators and plant and equipment. Proven construction plant and equipment is capable of enhancement, by the attachment of sensors and navigational aids, to provide improved feedback to the operative. Under some conditions, productivity can be increased dramatically. In one striking example, a (earth-moving) grader can increase its output by a factor of four. The implications of this are many, not least the difficulty of other operators in the same marketplace being able to compete.
Work at the University of Lancaster has produced a prototype driver-less excavator, called LUCIE, based on a small excavator. The addition of sensors and controls to enable program-controlled operation means that digging and placing of spoil is accomplished automatically once the machine is positioned in front of its work area. We might be tempted to place this development in the third category below, that of the intelligent machine, but will defer for the time being. At some point, we would expect to promote it to that category.
There are other machines which are capable of demonstrating significant improvements over entirely manually-controlled methods. The use of laser controls is the most common, though ultrasound has its place too. In one application area, that of large pour concrete screeding, laser-controlled machines have come to the fore in just a few years. This has transformed what had been a low technology area into one that has raised productivity and lowered costs significantly. UK specialist concrete flooring contractors are now able to perform at the highest levels of anywhere in the world. If there is an area for improvement, it is in the supply of consistent quality concrete. Once at the forefront of construction automation, concrete batching and delivery (ready-mixed) is increasingly coming under scrutiny for its variable quality performance.
- Task-specific, dedicated robots
Probably, the most publicised examples of these are those developed by the Japanese Big Five. To be fair, other Japanese construction companies have developed working construction robots. They include Kumagai-Gumi, Hazama, Fujita and Tokyu. There are many examples: concrete placing, levelling, screeding and power-floating; steelwork lifting and positioning; spray-applied fire-proofing; and rebar placement. It would be wrong to give the impression that any construction robot would have to be Japanese, though they have been the most active by a long way. Later, we shall consider the efforts of other countries in this regard.
Task-specific, dedicated robots generally work under teleoperation or program control. The operative has been removed from the immediate vicinity of the machine and instructions are transmitted from a pendant controller. Depending on the configuration of the machine, an umbilical link may be used to supply power as well as transmit control signals. The robot performs a well-defined task and has been shown to produce productivity savings of a worthwhile order. Adaptation to other tasks is generally not possible.
- Intelligent (or cognitive) machines
There are very few examples in this category, but that is not to say they will not appear before too long. LUCIE falls potentially into this category, but requires a decision-making capability to enable it to position itself in front of the work area and to determine its own work plan. As machines of this kind appear, it is likely that they will have developed from both of the categories above and, in this sense, they will represent convergence of the technologies. It is possible, but by no means probable, that technology from space exploration will eventually spin-off into construction. Of all the hostile environments - nuclear and marine included - space represents the ultimate challenge for developers and one where construction is to be found. These hybrid forms of robot will be distinctively construction-orientated, having a high degree of autonomy and knowledge-base with which to resolve the wide range of problems found on sites. It is more than a coincidence that Shimizu in Japan has maintained a section in its Institute of Technology dedicated to construction in space.
Solve those kinds of problems and more terrestrial activities might become almost trivial exercises!
- Achievements and progress on a global basis
There is a distinctively geographical dimension to developments in construction robots and automation that we should attempt to classify or group according to countries or regions. First, we shall consider Japan.
Much has been said already about R&D and the short-run production of construction robots which are now being used on sites. The greatest concentration of effort is to be found in the construction companies, with some government-funded agency work and complementary developments within the universities. The latter is, however, relatively small-scale. That said, Japanese construction companies are increasingly working in collaboration with American universities (Purdue, Stanford and Austin, Texas) and is likely elsewhere, particularly in the UK. One striking feature of Japanese efforts is that there appears to be significant duplication, with each of the major players having developed its own robots. The likely reason for this is that each has had both the capacity to innovate as well as being expected to do so by its customers. Being seen to be innovative is important in Japan.
Second, North America is host to a wide range of developments and initiatives: some industry-based, many university-based. The pattern is one of cooperation across a broad front, where academics, researchers and practitioners are brought together. Pure industry-based work is far less evident than in Japan. Nevertheless, Bechtel and Brown & Root have been active, alongside the University of Austin, Texas. Other universities include those mentioned above in the Japanese connection, together with MIT and Carnegie-Mellon. Example developments include pipe manipulation, pavement cracking inspection, stud welding and block laying.
Third, European efforts have been somewhat hit and miss. In Finland, VTT (Technical Research Centre) has worked steadily on several prototype devices though has found industry participation and take-up wanting. In France, CSTB's efforts are more a matter of record than activity in masonry laying and spray-applied finishes. Enhancements to plant and equipment used in concreting have been the mainstay of German efforts; whilst in Israel at the Technion, several prototypes of the kind seen elsewhere around the world are to be found: masonry laying, trussed rafter production and spray-applied finishes. Finally, in the UK, most of the effort over the past five years or so has been in the universities, with Reading (design for automation), Imperial College (simulation of jointing), City (masonry laying), Lancaster (excavation), Portsmouth (wall climbing) and the West of England (wall climbing) active to varying degrees.
Last, the Russians have worked largely in the area of heavy plant and on program control, though have lacked the computer hardware with which to enable major progress to made. There are other examples around the world including some in Australia.
- Ways forward
- Construction robots
In the mid 1980s, a sensible prediction might have been to believe that by the end of the century, robots would be commonplace on construction sites, certainly in Japan. The pace of development was such that it seemed just a matter of time. By the late 1980s, it became apparent that a plateau had been reached. The easy tasks, from an automation perspective, had been tackled, but there remained too many problems to solve before widespread use of robots could be expected. Moreover, there was a growing realisation that the nature of design and production would have to change significantly. Consequently, there was a move towards re-engineering the construction process, beginning with rationalisation of designs and the adoption of design for automation principles. Yet, some development of task-specific robots has continued and, from the evidence at hand, this will continue though with a difference. Instead of pursuing development to the point where a critical mass is reached, in terms of robots of sufficient number and concentration to automate the production process, a new industrial paradigm has been adopted. The advent of the automated site - a kind of factory that builds itself - will form the centre-piece of development well into the next century. Task-specific robots will be developed within a framework that emphasises total control over production and quality performance.
Before we leave this subject, it might be useful to summarise where effort will lay in the coming years in task-specific robots and the problems that have to be solved. Essentially, there are four areas which need attention:
- navigation
- sensing (especially vision)
- rationalisation and standardisation of components
- investment for R&D.
Assuming that progress continues along the lines of that experienced over the past five or so years, we can detect three distinct paths:
- enhancement of existing plant and equipment using low-cost technology to boost productivity and effectiveness
- task-specific, dedicated, present generation robots executing discrete tasks, but operating on simplified building technology
- intelligent or advanced robots that can execute complex, ill-structured tasks.
Accompanying these developments will be changes in several areas:
- construction processes will move towards simplified assembly and fixing
- off-site prefabrication will continue to increase
- engineering services and fitting out will be more closely coupled
- operatives will be retrained in technology-based skills' areas and act more like generalists than crafts-based specialists
- integration of design and production though the closer working of designers, constructors and specialists.
Activity 4
Try to assess the relative importance of each of the above changes to your company (or one you know of) by ranking them in order of priority. Annotate each with the short and medium term actions which you feel should be adopted.
Box 1: Rank order; Box 2: Short term actions; Box 3: Medium term actions
- Total automation
So far, we have only touched on the potential for the totally automated construction site. In this part, we describe briefly the background to major initiatives and then select one for closer examination. Important in this respect is that we can now appreciate why developments have moved in this particular direction: the previous sections were intended to serve that purpose.
Recognition by at least three of the Big Five Japanese construction companies (Shimizu, Taisei and Obayashi) that a revolutionary approach to site automation was needed is manifest in the systems of SMART (Shimizu), T-UP (Taisei) and ABC (Obayashi). By redesigning the entire construction process and using computer simulation, it has been possible to recreate the working environment of the factory as a construction site. Of the three systems mentioned, SMART has the highest profile and has moved beyond prototyping. Completion of the Juroku Bank office building in Nagoya in late 1993 has been followed by a second project in Yokohama.
- Case Study: SMART (Shimizu Manufacturing system by Advanced Robotics Technology)
SMART represents more recent attempts at computer integrated construction (CIC) - discussed below - that claims to reduce by 30% the number of man-hours required to complete a multi-storey office building (Normile, 1993). System set-up takes about six weeks, after which the building's top floor and roof are erected on top of four jacking towers: the effect is to resemble a top-hat. The jacking towers are used to push up the 1,323 ton top floor assembly - the main work platform - as well as lifting their own bases from floor to floor in a cycle time of around two and a half hours. The heart of the system is composed of lifting mechanisms and automatic conveying equipment which is installed on the work platform. This later becomes the roof of the building. Overhead gantry cranes are connected to the underside of the roof structure in a way that resembles a factory production facility. Trolley hoists are used to lift and position components which are introduced at ground level.
The whole process is computer-controlled, though workers are still involved in overseeing operations at least for the time being. Simplified connections between components facilitate rapid erection times: self-centering column connections require only fine-tuning with a torque wrench and a laser-guided gauge. A clamp-on welding robot - one of a few task-specific devices - is used afterwards to effect the final mating of the column ends. Floors emerge from under the top-hat pre-clad - again from the inside - allowing work in fitting out to begin immediately. Weather is excluded from the job-site by a mesh fabric hung around the work area. Racks of pre-assembled pipework are a further example of an entire approach to rationalising design and production, the aim of which is to drive down the man-hours required for production.
SMART automates a range of production processes including:
- erection and welding of steel frames
- placement of precast concrete floor planks
- exterior and interior wall panels
- installation of various prefabricated units.
Inevitably, with the first run of anything, costs are higher than normal: it would be unusual if it were not so. However, further improvements to the system will enable it to have wider application and, hence, lower costs because of economies of scale and familiarity with the technology. Likewise, lower cost will make the system more attractive.
- Computer integrated construction
SMART serves many purposes. Apart from the obvious gains as outlined above, it forms part of a much broader strategy for construction. CIC or computer integrated construction is the concept within which SMART is placed (Miyatake, 1993). In this sense, SMART is being used to demonstrate an approach which aims to integrate the entire AEC (architecture/engineering/construction) process. It achieves this by bringing together three elements:
- integrated design and construction planning
- a site automation system
- factory automation.
Innovations in IT, typically KBE, database management, simulation, engineering and management software, 3-D CAD and object-oriented programming have opened up new possibilities for systems' integrators. SMART is, thus, a prime example where this has been achieved.
CIM or computer integrated manufacturing has been used as the model for the site automation system element of SMART. Technologies such as just-in-time (JIT), materials' handling, process control and inventory control are implicit in the approach. The third element, site automation, brings together a raft of technologies and management practices that are adapted to the circumstances of the construction site. Automated transportation of materials, followed by their assembly and positioning using robots completes the process.
Performance targets for CIC projects are demanding. Shimizu has set the following, seeing them as entirely realistic:
- reduction in the total time of the project 50%
- reduction in the total manpower requirement 50%.
The implications of the push towards total automation, as embodied by CIC, is to require many changes to working practices. For Shimizu, this means that job descriptions for its workers have had to become less specific because of the integration of different functions. The emphasis has moved the company still closer to multi-disciplinary working. Within this, Shimizu has recognised that, not only are technical skills important, so are the personalities of the workers.
- On the R&D agenda
R&D continues to explore many areas. The following is a list of current work topics:
- construction robot control
- construction robot design techniques
- information systems and technologies
- robot safety and safeguarding
- sensors and sensing techniques
- user interfaces
- automated planning and scheduling
- bridges and road work
- building construction
- building finishing
- meteorology in construction
- special applications.
- Future form of construction automation
If SMART represents the extent of current achievement, what will be the next step in the quest for fully automated construction? For the answer to this question we should look at the IF7 project in Japan. Aptly named the Intelligent Field Factory , the aim is to bring advanced (i.e. intelligent) manufacturing to construction. The promoters of the project, Hitachi Zosen, Kajima, Shimizu, Hazama, Waseda University and MITI, are likely to be joined by researchers and industrialists from Europe and Canada under the support of the IMS (Intelligent Manufacturing Systems) initiative. Conceptually, the project is simple enough: it will build upon the success of SMART and systems like it to deliver the intelligent site of the future.
The concept operates on two levels. First, there is the real construction world in which people and robots co-work and interact with one another. The second is cyberspace within which is held a vast amount of information to be used to help deliver buildings or other large volume structures anywhere in the world. In this respect, it is important to note that buildings constitute just one type of structure to be encompassed by the IF7 project; another type is ships.
In the real construction world, large volume structures would be built largely with pre-assembled components. Intelligent robots would work to assemble predominantly heavy components which would be complete with knowledge of how they are to be processed and so on. Each robot would be able to communicate easily with people. Cyberagents which exist in cyberspace will be used to support decision-making in the real construction world.
At the end of this section you will find a personal feedback section. The answers can be
found once you have completed the exercise.
- References and bibliography
- References
- Normile, D., (1993). Building-by-numbers in Japan. Engineering News Record. March 1. 22-24
- Miyatake, Y., Yamazaki, Y. and Kangari, R., (1994). The SMART System Project: A Strategy for Management of Information and Automation Technology in Computer Integrated Construction. Proceedings of 10th International Symposium on Automation and Robotics in Construction. Brighton.
- Recommended bibliography
- Proceedings of the International Symposia on Automation and Robotics in Construction, held annually and organised by the International Association for Automation & Robotics in Construction.
- Automation in Construction, published by Elsevier Science BV.
- Personal feedback questions
PF1. Why should an incremental approach to the introduction of construction robots appear not to work, given the opportunity to reduce costs directly through improved productivity?
PF2. What is arguably the greatest inhibitor to the introduction of construction robots and automation in most countries?
PF3. How can construction companies develop meaningful strategies for embarking upon programmes of R&D which would have the capacity to deliver real gains in productivity and quality?
PF4. To what extent do you consider that the Big Five Japanese construction companies have, if at all, expended time and resources in developing much the same kinds of construction robot, only to have put them to one side in the face of SMART and similar systems?
PF5. Why should SMART and similar systems be of special interest when studying construction automation?
PF6. Is international collaboration in the work of the IMS, as applied to the concept of the intelligent field factory, the next logical step or is there something else?
You can find the answers to these questions by clicking on the button below.
|
|
|
