2005 Winch Kelsey What Do Construction Planners Do

Construction project planning is receiving growing attention as the limitations of formal deterministic planning are becoming more widely recognised.

In particular, the last planner and critical chain approaches are di? using rapidly. However, little of this debate has been informed by empirical examination of what construction project planners actually do. The research reported here draws on three di? erent research projects. One on the overall context of construction project planning, and two focused on requirements capture for the virtual construction site system.For the later project, 18 construction planners were interviewed on their daily practice. The results show that construction planning for principal contractors is more about negotiation with other interested parties and rapid decision-making based on heuristics than detailed analysis.

O 2004 Elsevier Ltd and IPMA. All rights reserved. Keywords: Construction project planning: Critical path analysis: Tendering 1. Introduction There has been considerable debate over the last decade or so on the e? ctiveness of construction project planning. However, there is remarkably little research into what construction project planners actually do.

The research reported here was undertaken as part of the requirements capture process for the development of the VIRCON system (VCS) as a strategic decision support tool for construction project planners. This larger research project is reported elsewhere [1]. This paper will explore the evidence on the role and practice of project planners that underpinned the development of the VCS.In so doing, it will address what Laufer and his colleagues [2] identify as the operation/systems analysis level of construction scheduling. The paper will ? rst review some of the recent debates on construction project planning. It will then report the results from our inter* views with 18 construction planners from ? ve leading UK ? rms before some conclusions are drawn.

2. Recent debates in construction project planning In an early paper, Laufer and Tucker [3] provide a critique of (US) construction planning.They argue that: The planning and evaluation of planning processes are non-existent. There is over-emphasis on critical path methods. Planners lack construction experience.

Planners have poor information gathering methods. Planning is control-oriented instead of actionoriented. Plans are poorly presented with overly-complex information. In a subsequent paper, Laufer and his colleagues [4] look at the de? nition and allocation of planning work. Corresponding author. E-mail address: g.

winch@umist. ac. uk (G. M. Winch).

0263-7863/$30. 0 O 2004 Elsevier Ltd and IPMA. All rights reserved. doi:10. 1016/j.

ijproman. 2004. 06. 002 142 G. M.

Winch, J. Kelsey / International Journal of Project Management 23 (2005) 141–149 They found that there was no clear system used, and planning was done in a multiplicity of ways. Laufer and Tucker [5] discuss the problem of who should do construction planning and when they should do it. The specialist planner has the time to do the work but incomplete practical knowledge. The line manager has the practical knowledge but does not have the quality time to carry out the task.

The specialist planner has better strategic decision-making skills than the short-term decision-making focus of the line manager. Line managers see the delegation of key decision-making to another as a threat to their position. These problems are also con? rmed in a wider project management study [6] and can result in: The planner preparing a plan which has incomplete information and inadequate decision-making authority. The line manager treating such plans as merely an irrelevant forecast prepared by another. For the whole project process, the ability to in? ence cost diminishes rapidly over time, the greatest in? uence being at the design phase. However the greater the time between plan and implementation, the greater is the variance of actual schedule and budget against plan.

Detailed planning of activities planned to be executed far into the future – say more than three months – is futile. This is because: there are uncertainties which cannot be quanti? ed; stochastic modelling does not identify the cause of variation; forecasting models extrapolate from past trends which may not be valid for the future.Theoretically, the binomial theorem [7,8] supports this claim. The probability of an event occurring in repeated trials – where the event may be, for instance, a task running 10% over its planned duration – is given by the binomial distribution. This shows that the percentage error of the estimate of the mean performance on the task as a ratio of the percentage error for the mean durations of all tasks in the work breakdown structure (WBS) is inversely proportional to the square root of the number of tasks, for a discrete sequence of tasks.

Thus for a 52 week project to be estimated at ±10% accuracy for handover date, a task with a duration of 2 weeks should only be planned with a ±50% accuracy, because some tasks will exceed their planned duration and others will not require it. This calculation, therefore, de? nes the lowest planning unit for the WBS. The larger the number of tasks in the WBS, the less point there is in planning each task accurately, because positive and negative variances will even themselves out at the level of the project as a whole. This argument, f course, does not apply where there are systematic biases in estimating. 3.

Responses to the critiques of construction project planning These critiques have stimulated a number of di? erent responses. Goldratt [9] argued that there has been nothing new in project management in 40 years, and advocates the critical chain method (CCM). This addresses two of the key problems of CPM – the inherent uncertainty of task durations and the associated opportunistic behaviour in establishing the true duration of tasks, and the resourcing of tasks.In CCM, the critical chain is the longest resource constrained path through the network, theorised as a constraint to be elevated. Thus a critical chain looks like a critical path, but it includes resourcing in the dependencies.

The conceptual shift from critical path to critical chain by including resourcing issues in the latter might be considered simply a technical development, moving on from the resource levelling approaches which are well established. However, the elevation of the constraint introduces a much more radical aspect.This elevation starts from two observations: That actual duration for any planned task is unknown, but can be assumed to be distributed as a gamma distribution with the median task duration greater than the mode. Existing estimates of expected task durations are padded because they include safety time, greatly extending the length of the critical path. In a situation where actual task duration is uncertain, and managers are held accountable for meeting stated durations (deadlines), then task duration estimates are going to be at the worst case end of the distribution, otherwise managers risk overrunning their deadlines.

CCM instead proposes that the average estimated duration should be used. An inevitable result of this is that managers will overrun their planned durations half the time. This problem is solved by removing from managers the absolute responsibility to meet task-level duration deadlines, providing incentives for early completion instead. This approach was tried on the A13 road project in east London with great success [10].As a result of these considerations, something of a consensus is emerging that construction planning has to be a much more decentralised activity in order to cope with the inherently uncertain nature of task duration. This issue is addressed directly by the developers of the ‘‘last planner” approach [11] who argue that construction planners should make only ‘‘quality assignments” where tasks: G.

M. Winch, J. Kelsey / International Journal of Project Management 23 (2005) 141–149 143 are su? ciently well de? ned, – to be co-ordinated with other work and – for the inputs to be identi? d and assembled, are ready to start because, – all materials are available, – design is complete, – precedent works are complete, have priority in the critical path for delivery to the customer, are commensurate in scale with the available labour for the coming week, are carried out within a system where the causes of incomplete or poor quality assignments are investigated and identi? ed. They argue that tasks not meeting these criteria should be deferred (even if this means a short-term loss of output against schedule) and analyse the bene? s from following this course of action which they term as OshieldingO production. Despite the fundamental important of these debates for construction project management, most recent research has gone into the re? nement of existing planning tools, rather than research into exploring the organisational context of construction planning.

Exceptions to this generalisation include the A13 and study [10] and Laufer [6] who investigated involvement in the planning process, and found that it is an essentially collaborative process involving line managers, designers, and planners at various stages of the process.Faniran and his colleagues [12] investigated the optimum allocation of resources to the planning process, concluding that both too little planning and too much planning were factors leading to poor project performance. The aim of this paper, is to contribute to this limited empirical literature on planning practice. It will do this in two stages. First, the overall context of construction project planning will be explored through a single project case study which shows how the plannerOs work is located within a hierarchy of plans on the project.Second, data from interviews with 18 experienced planners will be reported in order to deepen our understanding of contemporary planning practice in the UK construction industry.

The research reported in this paper will, therefore, help us to ? ll in some of these gaps in our current knowledge on construction planning practice. 4. The context of construction project planning Fig. 1 is developed from a case study of a major hospital project in west London using a construction management procurement route, and shows the overall construction planning process.Italicized boxes represent contractually binding schedules. The research was ? nanced by the (then) Science and Engineering Research Council.

The clientOs project manager develops the clientOs strategic programme, which drives the tender programme for the procurement of the resource bases such as design consultants and construction manager, and also the architectOs design programme. The contractually binding agreement between the client and construction manager is the master programme.The construction managerOs project managers then prepare the target construction programme which guides the procurement of trade packages, and the parcel documents programme which drives the production of drawings by the architect. This last schedule is non-contractual, A tender programme B client’s strategic programme architect’s design programme summary programme master programme C target construction programme parcel documents programme tender restraints programme procurement programme D works contractor’s programme ontract restraints programme Fig. 1. The construction project planning hierarchy.

Source: Developed from: [8] Figure 11. 3. 144 G. M. Winch, J.

Kelsey / International Journal of Project Management 23 (2005) 141–149 but can be used in claims for delays against the client caused by the non-delivery of drawings. Within the target programme, trade contractors are given ‘‘windows” for the execution of their responsibilities on site in the tender restraints programme, and these are formally agreed in the works contractorsO programmes.Within those, the trade contractors then schedule task execution at the level of work breakdown structure (WBS) that suits them which is communicated to the construction manager in the contract restraints programme. In order to gain more bargaining power, construction managers may not reveal the master programme to the trade contractors, and provide a target construction programme that is signi? cantly tighter than the master programme, so as to bu? er the completion date to ensure a satis? ed client even if the works contractors programme slips (interviews 22/05/89; 30/09/89). 5.

What do planners do? urrent practice in construction project planning In order to explore further the reality of construction project planning semi-structured interviews were conducted between July and December 2000 with 18 experienced planners from ? ve leading UK construction ? rms. Interviews came from the principal contracting and trade contracting divisions of those companies. The two packages covered by the trade contracting subsidiaries were piling and mechanical and electrical services. The research was funded by two EPSRC awards – GR/ N/02917 and GR/R/51452 – and is more fully reported in [13].Most of the planners interviewed were currently involved in planning at the pre-tender stage including assembly and presentation of the tender documentation – in other words, activity A in Fig.

1. About half of those were also involved at the post-tender pre-construction stage – in other words, activity C in Fig. 1. Only a few were further involved during the site works. Generally, although a number of the interviewed planners had on-site experience the typical pattern is that a planner works either at the pre-tender, pre-execution stages or on site but not simultaneously.The exceptions tend to be where planners work for some time on a single large project.

In such cases it may make sense (from the employerOs point of view) for their work to carry on to the execution stage. For traditional contracting by single stage tendering the period for the preparation of the construction plan was around 4–6 weeks for larger contracts and 3–4 weeks for smaller ones. The post-tender to start on site period showed somewhat greater variation from 2 to 13 weeks. For two stage tendering, the ? rst stage was similar to the single stage tender period but it was only at the subsequent stage that a price had to be presented.However, for the planners the time frame was similar to the traditional method.

Those involved in two-stage planning reported that the plan was a signi? cant factor in progressing beyond the ? rst stage as a demonstration of competence was the paramount selection criterion at this stage. Under construction management procurement, planners tended to be brought in earlier in the process. The principal contractors were also involved in partnered contracts and PFI schemes where the tender periods were considerably longer (3–6 months and more).On large civil engineering projects, the tender periods were also longer, averaging around three months. However, planners often work simultaneously on several tenders, and so the actual working time which is available to them for preparing each tender submission is substantially less than the average tender period. Planners tended to be overwhelmed with information.

They received large amounts of information that was not relevant to their role, and spent considerable amounts of time sifting through it. As might be expected, this problem particularly a? cted the planners working for trade contractors. On the other hand, the quality of much of the information received was poor. For instance, M ; E drawings sometimes did not allow for any secondary steel support work, or installation space for pipework. Similarly, some designs showed a lack of understanding of the space required by piling plant, the most common problem being the siting of works too near con? ned boundaries making it impossible for plant to access the working area with su? cient working room. The methods for dealing with the uncertainty caused by design information de? iencies were: Guess the missing information based on experience and past job records.

Qualify the submitted tender. Assess the risk posed by the missing information and adjust the risk premium accordingly. Take a strict contractual stance on site with regards to negotiating the cost of variations to the tender drawings/speci? cations/scope of works. Table 1 presents the documents produced by the planner as part of the tender team. The ? rst four items were speci? cally mentioned by the interviewer, while the others were spontaneously mentioned by the informant.

Those items only mentioned by one planner have been omitted as being speci? c to that plannerOs type of work. One informant reported that for partnered contracts, he was prepared to list all the risk items in the tender with prices attached – a sort of risk Oshopping listO. It was then up to the client to decide which risk items it wanted to take on board. G. M.

Winch, J. Kelsey / International Journal of Project Management 23 (2005) 141–149 Table 1 Construction plannersO deliverable outputs Number of mentions Internal only Non-? ancial information produced at tender stagementioned by more than one planner Items mentioned by the interviewer Method statement Programme bar chart Critical path analysis 3 Risk assessment 7 Additional items mentioned by the planners Preliminaries scheme/resources Site mobilisation/layout drawings Design programme (for D ; B) Design information requirement dates Procurement programme Team details/cvOs/organigrammes Record/references of previous experience Buildability/programme/value eng. ptions 1 Quality assurance plans and procedure Health and safety plans and procedure Environmental protection procedures Phased work location drawings 8 Overall resource schedules External partnered only 145 External 1 18 18 8 3 5 5 6 5 6 7 3 3 6 7 3 4 3 Table 2 Domain- speci? knowledge acquired Type of learning mentioned by more than one planner Better understanding of site processes and works contractors Better understanding of M ; E services and co-ordination with other trades Development of better communication skills (including listening) Experience through working on a wide range of projects Better understanding of contracts and tender processes Development of the ability to anticipate problems Better understanding of the 3D/spatial aspects of the work (piling planners) Development of a feel for task outputs and durations Better understanding of supply chain management No 7 6 4 3 3 2 2 2 2Table 2 reports the domain speci? c knowledge planners believe that they have acquired which enables them to solve planning problems better than inexperienced planners. Older planners were critical of (some) younger planners. One felt that two generations of planners (taking a generation at 15 years) had now appeared who had little site experience.

The ? rst generation of OunsitedO planners was now teaching the second. Another felt that the younger planners were too technology driven instead of concentrating on developing a basic understanding of construction problems.Fig. 2 shows the preferred planning software related to the value of recent contracts handled by each planner. It is clear that Primavera is preferred for larger value contracts and Power Project for smaller ones.

A thorough review of the capabilities of these and other Critical Path Analysis software packages can be found in [14]. The Primavera users particularly praised its rigorous and disciplined task logic. The Power Project users criticised it for the same reason! The problem is that Primavera requires a substantial set-up time and considerable e? rt in maintenance and amendment. Amending one task in Primavera may force detailed adjustments of other tasks dependent on the amended task. Power Project will allow the last minute input of an OillogicalO task sequence which the planner believes will actually work but for which they do not have the time to re-input a new set of task dependencies.

This is vital due to the very short tender periods identi? ed above. According to our informants who use it, Power Project is easy to learn and easy to use with excellent presentation facilities.Users of both packages agreed that Primavera does require a signi? cant learning period. The informants who preferred Primavera software to Power Project also tended to make intensive use of 146 G. M. Winch, J.

Kelsey / International Journal of Project Management 23 (2005) 141–149 Average value of last five contracts (? m) 140 120 Primavera / Power Project 160 None / Power Project Power Project Power Project 80 60 40 20 0 0 Power Project Primavera 100 None / Excel Power Project / Excel Power Project Primavera 1 2 3 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Rank by average value of last five contracts Fig. 2. Decision support tools used by construction planners. Source: [13]. critical path analysis, as opposed to restricting themselves to Gantt charts. 6.

The construction planning process The approach to process mapping deployed for this research is a ‘‘business” one derived from the work of Rummler and Brache [15], rather than an ‘‘engineering” one deploying IDEF0 (see [16] for this distinction). Table 3 categorises the four types of actor in the construction planning process.Clearly the client and consultants initiate the process through the invitation to tender. The suppliers and constrainers determine the ability of the principal contractor to deliver the clientOs requirements, the former through their capacity and the latter by the extent of their agreement to allow the contractor to use the most e? cient construction processes where these impinge on their particular domain. The principal contractorOs planning team mediates between these other actors to form a viable plan.

The interaction with these actors also reveals a potential con? ict between short and long-term goals for planners. While they have to negotiate OrobustlyO with these actors on each contract, they also have to maintain a longer working relationship which may serve to moderate their negotiating behaviour. The key di? erence between the situation for principal contractors and works contractors lies in those actors in the OconstrainerO category, because they include other works contractors carrying out works precedent or concurrent in time or space.Whereas for the principal contractor, constrainers are known, for the works contractor these may be unknown at the time of tendering (depending on the construction procurement method) and therefore no negotiation with them is possible. This makes it all the more important that the principal contractorOs planner allows a su? cient distribution of space and other site resources to allow each works contractor to do their work unhindered. Fig.

shows the overall pre-contract construction planning process for a principal contractor highlighting: The interdependency between construction site programme planning and design/procurement programme planning. Table 3 Principal contractorsO pre-tender construction planning process actors Role Customer Description Ordering built product, representing clientOs interests and supplying design information not controlled by the contractor Planning construction as part of tender o? r to supply built product Providing contractor with information on supply chain delivery capabilities Controllers of the site and its environment who have the power to block or restrict construction operations Actors Client, project manager, architect, principal quantity surveyor, structural and M;E engineers Principal contractorOs planning team responsible to senior management Trusted works contractors and specialist material suppliers Client and/or clientOs landlord utilities, adjoining owners, police, regulatory/highway authorities Performer Supplier Constrainer G. M. Winch, J.Kelsey / International Journal of Project Management 23 (2005) 141–149 147 Consultant Total Contractor’s STAGE Construct’n Design Design Procurem’t Design Programme Information Information Programme Droduction Planning required required ACTOR Programme dates dates Consultant Principal Contractor Package Contractor Material Supplier Constrainers Key Decision Action Consultation Fig. 3.

The principal contractorOs pre-contract construction planning process. Source: [13]. The general sequence for determining the dates when design information is required from actors outside the principal contractorOs supply chain. The substantial involvement of supply chain actors. The important decision role of constrainer actors.

The sequence of action which is being planned is clearly of the general form Design ! Procure ! Construct However the actual programme planning process has to be in reverse – or ‘‘backtimed” – due to the driving role in planning of the clientOs strategic programme, box B in Fig. 1 as follows: Construct ! Procure ! Design In a very important sense, from a planning perspective, design and procurement are merely inputs into the execution process on site which delivers the asset to the client ready for exploitation.This can be illustrated by the following sequence of programme planning, as identi? ed in the horizontal axis of Fig. 3. Stage 1 – identi? cation of the overall site construction programme 1. 1 Breakdown of identi? ed construction processes into packages according to supply chain capacities.

1. 2 Planning of sequential logic of processes/packages in time and space. 1. 3 Assessment of common site resources and temporary/enabling works required to service direct construction processes. 1.

4 Negotiation with constrainers (as de? ned in Table 3) to allow certain construction operations. . 5 Informal negotiation with supply chain representatives (and/or experienced in-house operations manager) to con? rm or amend steps 1. 1–1. 4 Stage 2 – identi? cation of the overall procurement programme 2. 1 Identi? cation of all key supply chain actors including those excluded from on-site construction processes.

2. 2 Identi? cation of lead times for key supply items and works contractor tendering. 2. 3 Planning of sequential logic of procurement to support construction programme. 2. 4 Informal negotiation with supply chain representatives to con? rm or amend steps 2.

1–2. 3.Stage 3 – identi? cation of design information delivery dates required for procurement and construction 3. 1 Identify all detailed design requirement dates (from all sources) to support construction and procurement processes Stage 4 – identi? cation of the contractorOs design programme 4. 1 Identi? cation of temporary works design requirements. 4.

2 Identi? cation of supply chain design responsibilities. 4. 3 Identi? cation of supply chain information requirements to carry out design activities. 148 G. M.

Winch, J. Kelsey / International Journal of Project Management 23 (2005) 141–149 . 4 Identi? cation of clientOs consultantsO lead times required to approve design data for which the contractor is responsible. 4. 5 Planning of sequential logic of contractorOs design activities to support Stage 3 above.

Stage 5 – identi? cation of the detailed design data requirement dates from clientOs consultants 5. 1 From Stages 3 and 4 above computation of all outline/detailed design requirement dates in order to support contractorOs design, procurement and construction programmes (the output of which may require repetition of Stages 1–4 above) The output from stages 1. , 2. 3, 4. 5 and 5.

1 will form part of the tender submission. As indicated under Stage 5, this is an iterative process and may require several passes before a viable integrated programme can be produced. Up to detailed design stage the client and the consultants negotiate both with each other and also with the local planning authority to arrive at an acceptable design. In this activity local regulatory authorities have to take note of objections by their own o? cers, the public and adjoining property owners at both outline and detailed construction permit application stage.At the construction planning stage there is a number of decision actors. This makes the plannerOs negotiating skills all-important.

Although works contractors do not decide whether or not to tender for part of the works at the principal contractorOs pre-tender stage, if principal contractor planners do not have the tacit support of their supply chain at this stage (or con? dence that such support will be forthcoming) then they are running a considerable risk. There are further problems in negotiating with the OconstrainerO category of actors.If construction permit applications have been passed against the advice of local authority o? cers or against strong objections from neighbouring property owners then, subsequent negotiations with these actors at the construction programme planning stage may be all the more di? cult. 7. Review of construction project planning practice It was clear from the interviews that a systematic review of project planning (and project review in general) was either rare or non-existent.

A signi? ant number of informants said that they did not tend to refer to past job records as these were either non-existent or inadequate. Project feedback does, of course, occur (negative feedback in particular). However, without systematic feedback it is not clear how learning and improvement of the planning process can take place at an organisa- tional level. Total project review is necessary to establish the reasons for project execution shortcomings. However, the actors most able to contribute to this process are those with the least motivation to expose their own errors.

Informants from the principal contractors were almost united in their approach to trade package identi? cation. This view was that the packages should be split according to the known trade capacities of individual contractors. Principal contractorsO planners can only realistically plan down to the works package level (broken down further into the main package components) in the target construction programme shown at C in Fig. 1. It is the trade contractorsO planners in the works contractors programme shown in D in Fig. 1 who have to turn these packages into a WBS to ield a sequence of individual tasks through time.

This practice vitiates against the recommendations of those arguing for the ‘‘clustering” of trades into technologically de? ned subsystems – see Winch [8] chapter 7 for a review. Nearly all informants reported that, having selected the construction method, they disregard the construction period stated in the tender and work out what time they think it ought to take to carry out the works. That is their OoptimalO period in the sense that it allows for the minimum but robust time bu? rs to cope with risks attached to critical packages. In nearly all cases this tends to be longer than the stated contract period in the tender documents. They then see what time they can trim o? the OoptimalO period. This exercise helps them to evaluate the risks associated with attempting to construct the building within the time preferred by the client/consultant.

They often discuss this with estimators who have a knowledge of trade output rates. Although they will take note of this, many are sceptical about relying on such data.They prefer to use their own OfeelO for package times gained by experience. Principal contractor planners do use critical path methods but their work is more in the nature of planning co-ordinators. They use critical path methods in a broad brush approach since they recognise that the people best placed to plan work in detail are the works contractors. What is clear is that those informants who plan larger more complex projects over several months attach greater importance to their use of critical path methods.

One of those was a works contractor so he has to plan in detail.The remainder recognise the limitations of the methods but point out that they have to allocate fairly the contract time between di? erent works contractors and they have enough experience and understanding of task dependencies to develop a workable network model. This assists them in negotiating with the works contractors. It also helps them to understand where the greatest time-risk areas might lie. Virtually all planners said that they were not attempting to optimise the construction time but rather to get a G.

M. Winch, J. Kelsey / International Journal of Project Management 23 (2005) 141–149 49 realistic time which allowed a reasonable time bu? er to take account of known risk areas. Most said that they would be prepared to Owalk awayO from a contract if the construction period stated in the tender were wholly unreasonable by the submission of a non-conforming bid. Most were prepared to submit a non-conforming bid if they felt that they could o? er a lower price.

A few would be prepared to submit a non-conforming bid at a higher price and a shorter construction period where they believed that time was of the essence to the client. 8.Conclusions Construction project planners add value for the contracting organisation by ensuring that estimating and tendering are based on a robust understanding of the methods, time and space required to carry out the tasks for each building contract and the corresponding risks involved. Their output can be in? uential in winning tenders not based on purely ? nancial criteria. They have to perform both planning and other work in a time- and information-constrained environment which requires considerable use of heuristics based on judgement and experience-based learning.

Negotiation and communication are important in their work and for this reason many of the processes and task interdependencies in their work are iterative. Their longer-term objectives have to take account of potential long-term relationships with operational management colleagues and representatives of other organisations with whom they have to negotiate. Pre-tender construction planning has to be understood within the broader context shown in Fig. 1 that includes design and procurement planning.It is a highlevel planning activity, which has to incorporate the many lower level plans of works contractors.

It involves as much co-ordination of planning as direct planning although re-examination of the lower-level schedule may be required to con? rm the robustness of the higher-level plan. It is this process that sets the framework for the more detailed level of activity as advocated by the last planner approach and critical chain which are most appropriate for the development of the works contractorsO programmes shown as D in Fig. . The bene? ts of planning are seen by mid-level managers as producing bene? cial future states that would not have occurred and by top managers as preventing adverse future states that would otherwise have occurred. Accordingly the view of the bene? ts of planning varies within the organisation. The OpreventedO states can be di? cult to identify and measure in economic terms.

It is clear that planning has a crucial role to play in delivering construction projects consummately [3,12].It is also clear that detailed planning needs to be decentralised to the level responsible for the execution of the works [10–12]. This paper, has focused on the pre-tender planning process which forms the basis of the negotiation and contractual agreement of the master programme, and also sets the strategic context for the more detailed decentralised planning to post-contract. We have shown that planners work in a complex network of relationships spreading within and without the contracting organisation, they work to pressing deadlines which preclude detailed analysis, and under signi? ant levels of uncertainty. We encourage others to pursue this line of enquiry – planning remains central to construction project management, and its practice deserves more research attention than it currently receives.