One man's meat...

For years I have argued that construction should not be regarded as any different from other industries and that what works for the automotive industry, for example, can equally work for construction.

I have to admit that research has now helped me see very different dynamics at play. Measures that have produced breakthroughs in productivity in industries such as automotive can lead to dislocation in industries such as construction.

This is particularly true of logistics. My research project, Efficiency and Effectiveness in Construction Logistics, sponsored by CITB ConstructionSkills, found that logistics played a central role in ensuring that projects are delivered efficiently. However, the research also suggested that the introduction of "lean" processes in parts of the supply chain to improve performance at the logistics level could have a negative impact on the industry's effectiveness.

The structure of the processes in the automotive and the construction industries look remarkably similar. They can be described as an "A-type" system where chains of activities are integrated and converge to a single point at the top of the letter A. In the case of the automotive industry this point is a completed car, in construction it is a finished building.

But analysed more thoroughly, striking differences emerge. The automotive industry operates a "process-centric" system of repetition and stability. Construction, however, has less repetition and a lot less stability - it is "project-centric". Within these systems the duration of activities is a key difference.

In the process environment operations last for seconds or minutes. In the project-centric environment activities can last for days or weeks. One is essentially routine and simple, the other is complex.

Construction's complexity is determined by the number of variables at play - dependencies at the task level, the extent to which different activities need to be integrated, and the level of uncertainty in terms of when a task or activity is likely to be completed.

To serve such a complex system logistics needs to be responsive and to be responsive there must be sufficient capacity in the system to deal with the unexpected events that are endemic in the construction industry.

Attempts to streamline systems and become "leaner", however, have tended to remove this capacity from the system. In stable systems, this is no problem: capacity costs money, and when it is not needed it makes sense to cut it.

In construction, the approach is dangerous.

As well as factors related to the project-centric nature of construction there are other issues particular to construction that further increase uncertainty and instability. These include fragmentation within the industry, the nature of short-term transactional relationships, lack of teamwork among trade operatives, and poor communication.

The research

The research project looked at three areas of logistics: transportation, stockholding and on-site efficiency of labour. These areas were highlighted by the Strategic Forum for Construction in a report published in August 2005.

The overriding conclusion of my research was that the logistics system is not customer-focused enough. The customer in this case being the site manager responsible for co-ordinating logistics activities on site. Some of the measures introduced into the industry in the name of efficiency are production-focused rather customer-focused. So driving efficiency in one part of the supply chain is making it less responsive and less customer-focused where it matters - on site.

The research also found that in some areas practices that appeared to be at odds with "lean" practices in fact made practical sense and enabled the projects to proceed more effectively than would have otherwise been the case.

Transportation

The research explored a range of issues including use of vehicles, travel distances and waiting times for unloading.

Often there was a clear trade-off between efficiency and responsiveness. Take, for example, the kitchen manufacturer who will only confirm the week of the delivery and not the day of the delivery. This means the manufacturer has a high degree of flexibility to ensure that only full loads are despatched, which is good for transport efficiency but not so good for project effectiveness.

But there is still a lot of wasted capacity on lorries due to the fact that there appears to be very few opportunities for lorries to make pick-ups from sites. Even returnable pallets, stacked ready for collection, are often left behind following a delivery. There does not appear to be an incentive for the driver to pick them up. This may be due to the fact that a lot of transportation is subcontracted by the manufacturers and the arrangements do not include pick-ups.

On the issue of lorries waiting, my research identified that not all waiting is uneconomic.

The overriding conclusion of the research was that the logistics system is not customer focused enough

It is common for lorry drivers with full loads to arrive significantly earlier than official site opening times. This means they can avoid rush hour, and they may be able to count waiting time as rest time for the tachograph and the timing of the drop off can mean his return journey will also be outside rush hour.

Stockholding

Stocks play an important role in regulating the flow of materials and ensuring that materials are available when needed. The management of stockholding on site is complicated by the fact that as much as 70% of materials might be ordered by trade contractors.

There has been a general move by construction material suppliers to reduce stockholdings of finished goods to achieve cost savings and improved cashflow. What might have been standard stock items are now often offered on a made-to-order basis. The result is that lead times have increased and become more variable. Scheduling material procurement in advance certainly helps, but this is not always an option. The supply chain should be becoming more responsive, not less so.

There is also a trend to reduce stockholdings on site in favour of just-in-time deliveries. This is fine if material usage can be accurately predicted and we are working in a stable production environment. But given that management resources are generally pretty stretched, just-in-time deliveries could be just another issue to worry about.

Then there are times when things sometimes go better than planned. More labour suddenly becomes available and we can push ahead with the schedule. There are generally about nine conditions that have to be satisfied for a task to start (see box) and one is the availability of materials. Do we really wish to risk material availability being the constraining factor when we have so many other more problematic issues to contend with?

Another area under review in efficiency drives is double handling of stocks on site, which is seen to add costs and increase the risk of damage. However, the research highlighted the important role that carefully managed double handling can play. Rather than being a negative feature of stockholding, double handling can be seen as a positive part of the material handling process. Measures to store materials securely on site, and to protect them from damage, both before incorporation in the building and after, all require additional handling.

It is often a price worth paying to ensure defect free finishes.

On-site labour

Labour is the resource that is most critical in progressing any project and its efficient use is an important factor when looking at project effectiveness.

At the logistics level the principle disruptive factor is the availability of labour. This is most pronounced during the mid-to-latter stages of the project where there is more parallel working and greater dependency between trades for completing and starting tasks.

The delayed completion of tasks due to inadequate labour availability appeared to be a significant factor in disrupting projects. This led to interruptions in workflow of other trades with the consequent knock-on effects to the programme.

The key issue at an operational level is the lack of reliability of trade contractors in committing the necessary resources to undertake the agreed work.

Secrets of success?

The research showed that investment in site prelims - such as site office set up, plant and equipment and storage facilities - and supportive management practices are key to successful logistics management. Unfortunately, the current practice of procuring projects which focus on lowest site prelims, overhead and profit, is encouraging the opposite to happen.

Supportive management practices include the establishment of a logistics plan, scheduling material deliveries and crane usage and the development of supply chain arrangements.

One management system that recognises the complexity in the procurement process is buffer management. This system is particularly appropriate for materials subject to a lengthy procurement process involving design and manufacture before delivery and which are most likely to cause disruption to the project through failing to arrive at the right time. Examples include windows, external doors, bespoke joinery, sheet roofing and roof trusses.

The buffer management system identifies key dates in the procurement programme by which each stage of the process has to be completed, including any approvals and redesign. A time buffer is inserted before the delivery date to accommodate delays to any of the process stages.

By monitoring how much of the buffer is used, this system shows how the project is performing. A traffic light system in the form of a "Fever chart" provides a visual tool to signal the status of the procurement programme.