Building information modelling (BIM) – enabled construction education: teaching project cash flow concepts

Abstract

This research explores the practical feasibility and effectiveness of BIM-enabled education in teaching the topic of project cash flows to construction management students. Using a participatory action research methodology, a BIM-enabled cash flow exercise was developed, carried out and refined in a construction investment course to simulate integrated practice. The results of the implementation demonstrate that BIM-enabled education can promote and infuse both BIM collaboration and professional practice experiences within an architecture, engineering, construction, and facilities management (AEC-FM) curriculum. Additionally, the teaching practice and method in this intervention demonstrate the capability to accommodate all levels of knowledge in Bloom’s taxonomy which is a standard requirement for educational module design. This study recommends that BIM-enabled education be embraced and explored by faculties in AEC-FM courses to improve teaching and learning of construction management concepts.

Keywords:

• Building information modelling (BIM)
• AEC-FM education
• project cash flow
• construction investment
• construction management
• engineering education
• BIM exercise
• problem-based learning

Introduction

Innovations and improvements in didactics arising from the digitalisation of the construction industry are being continuously witnessed in Architecture, Engineering, Construction and Facilities Management (AEC-FM) programmes. Building Information Modelling (BIM) is a central feature of this digitalisation, and it presents multiple opportunities for teaching practice improvements. BIM provides a collaborative system of construction that enables the digital representation of physical and functional properties of construction assets with which stakeholders can interact and it ensures comprehensive, organised and readily accessible project data. Much of this data is referenced directly to building objects (walls, beams, columns, windows, doors, floor slabs, pipes, etc.) which are represented in a virtual, 3D model of the building so that they can be easily viewed and understood. BIM therefore enables the simulation of real and complicated project scenarios in the classroom in a way that is both efficient for students to grasp and which promotes students' familiarity with BIM workflows and technology. In addition, numerous studies have shown that learning can be enhanced and students' motivation increased by introducing technology into didactics (Barham et al. 2011a; Wu and Kaushik 2015; LóPez-Zaldívar et al. 2017; Shanbari and Issa 2019).

This research explores the practical feasibility and effectiveness of BIM-enabled education through an action research case study focused on the topic of project cash flows as taught to construction management students. It is motivated by the potential for a new, BIM-enabled educational approach that moves away from dividing projects between specialist areas and toward integrating work and information flows for whole projects (Forgues and Becerik-Gerber 2013; Shanbari and Issa 2019) and which embraces the opportunities of BIM as a learning environment (Witt and Kähkönen 2019; Zamora-Polo et al. 2019). This offers solutions to the perceived mismatch between graduate competencies and their professional roles in industry (Forsythe et al. 2013; Lim et al. 2015), the need to integrate students' learning in the context of real projects (Alshanbari and Issa 2014) and, in doing so, to promote experiential, student-centred learning methods such as problem-based learning, etc. (Becerik-Gerber et al. 2012; Park et al. 2016; Wu and Luo 2018).

Project cash flow as a learning topic for this intervention is based on the experience of the facilitators and its relevance in professional practice as an aspect of project cost management. In learning topic selection, Ahn et al. (2013) suggest that the learning topics should comprise and reflect a selection of knowledge, skills, values, and attitudes relevant to and valued by the profession, subject disciplines, and by the wider society. Although cash flow concepts are common in construction management education, the way they are taught is predominantly by the traditional method of talk and chalk (Mills and Treagust 2003). Mills and Treagust (2003) reveal the inefficiencies of this method and the calls for change by engineering accreditation bodies. BIM-enabled education is a way of answering this call for change, but the number of documented BIM-enabled education is sparse, and thus suggests that this mode of knowledge propagation is still in its infancy. Using an industry BIM workflow ensures that the scenario on which the learning activity (project cash flow management, in this case) takes place corresponds to industry reality and that the data input to the learning activity is not contrived by the lecturer but rather exists as real project data and is drawn directly from the same sources as would be the case in industry. For the topic of project cash flows, the appropriate data input (time and cost) relies on a relatively detailed (5D) BIM model, and this helps to emphasize the role of BIM in organising all project data while also associating that input data with the spatial/physical aspects of the construction project which students can easily relate to. This enables closer alignment between the taught topic and the contemporary, digitalized industry workflows which are emerging in the construction industry globally. As a result, the experiences acquired and recounted here, as well as the analysis and conclusions of this effort, can benefit a large number of institutions all over the world.

This study is an initial step in developing an innovative, BIM-enabled curriculum that integrates construction management concepts in experiential learning on the basis of real construction project data, which is now, with the digitalisation of construction, increasingly possible in a classroom setting. This paper describes a BIM-enabled education intervention developed through action research, its objectives, organization, development, and instructional approaches. It provides details of the learning outcomes and contextual evaluation of this new approach and summarizes the lessons learnt. In the next section, this research is set in the context of existing BIM-enabled education initiatives reported in the extant literature. This is followed by a description of the action research methodology applied and then of the case context in which it was applied. The findings at each stage of the action research process are then presented before a discussion of their implications for educational practice and, finally, conclusions are drawn.

Literature review

Construction education

The current educational system has functioned with reasonable success for decades, however, the needs of the current generation of students are distinct, necessitating a unique strategy in order to achieve similar or even better results. Instructors around the world are aware of this and have recommended other methods of conducting university classes (Hanford 2011). Using a driving lesson example, Alshanbari and Issa (2014) give a succinct picture of the relationship between the learner and the industry. They opine that it is difficult for an intending driver of a car to develop the required driving skills without getting behind the wheel and driving. The practical part is necessary to complement lectures and tests which in themselves are not enough to produce a good driver. Therefore, they suggested that higher education should operate in a similar manner by encouraging students to engage in discussion and hands-on experimentation (Alshanbari and Issa 2014). By this, they canvass for a shift from a purely behaviourist based educational approach to a more constructivist based educational approach.

According to Barfield et al. (1995) learners can be classified as auditory, visual, tactile, or kinesthetic based on their learning preferences. Auditory, visual, and kinesthetic learners learn through hearing, seeing, and doing respectively (Roark 1998). Teaching AEC-FM classes while considering students' various learning styles is a difficult task (Barham et al. 2011b). Barham et al. observe that traditional lecturing is prevalent in delivering AEC-FM courses with occasional visits to construction sites sometimes used to complement the lecturing approach. This teaching style provides an auditory and visual learning environment. However, site visits are not always possible to include in the course schedule due to factors such as the lack of construction sites that match the class's demands, class scheduling conflicts, and safety concerns (Haque 2007). The typical lecture teaching method can often fall short of serving as an effective instrument for communicating knowledge to learners (Barham et al. 2011b). AEC-FM students are now unable to develop the necessary abilities to tackle real-world problems due to the lack of a favourable learning environment that stimulates aural, visual, and tactile senses. To improve students' learning capacities, a user-friendly interactive information repository with a conducive learning environment is required.

Innovations in construction education

Since the launch of "Cyclone" in 1976, using computers to simulate building jobs has been the subject of numerous research articles (AbouRizk and Shi 1994). Researchers now have more resources to run more accurate simulations because of the advances in technology. Modelling a full construction project, on the other hand, is far more difficult than simulating one or two components of it separately. Construction projects include an excessive number of activities and unpredictable variables, such as weather conditions, that can radically alter the outcome. Furthermore, one task can influence and be influenced by the others. The intricacy of construction projects is one of the key reasons why proper simulation is so challenging. As a result, numerous researchers have turned their attention to simpler simulation for instructional purposes (Nikolic et al. 2011).

BIM makes it easier to create knowledge libraries and learning settings that are favorable to learning. For data management, BIM is a wonderful tool. Through the 3D spatial model of the construction object, it enables easy and quick access to information contained in a single centralized database or in multiple databases kept at separate places. Auditory, visual, and kinesthetic learning environments occur as a result of BIM qualities such as simple access to information, visualization, and simulation capabilities. Access to the repository at any time and in real time via a 3D model creates a learning environment that is free of time and space constraints, allowing students to learn at their own speed (Barham et al. 2011b).

BIM as an innovative concept in construction education – The evolution of BIM education

BIM emerged from origins in “Building Description System” through several incarnations (such as Building Product Model, Product Information Models, Virtual Display Model, etc.) between 1975 and 1992 (Eastman et al. 2011). Through the introduction of policies, standardization, and improved accessibility in the early and mid-2000s the construction industry has witnessed a surge in BIM adoption and this has generated the need for new skills and competency requirements for graduates and industry professionals (Hooper 2015; Govender et al. 2019). Construction education has struggled to keep pace with industry BIM innovations and this has created a gap between graduate competencies and their expected professional roles in the industry (Barison and Santos 2018; Bozoglu, 2016).

The evolution of BIM education has been conceptualized into 3-progressive stages (Underwood et al. 2013):

1. BIM-aware, where graduates are made aware of the uses and exigencies of BIM relating to its implications for both digital and cultural transformation of the construction industry.

2. BIM-focused, involves graduates’ abilities to use and manipulate BIM software in performing specific tasks such as modelling, clash detection, simulation etc.

3. BIM-enabled, where education takes place in a BIM-mediated virtual environment and BIM acts as a platform for learning (Underwood et al. 2013).

Evidence suggests that both BIM-aware and BIM-focused education have been generally recognized and initiatives to develop curricula to incorporate BIM have become widespread. For instance, Peterson et al. (2011) reported how they introduced BIM to students (BIM-aware) and subsequently used it to demonstrate project management techniques (e.g., line of balance), integrated design and design optimisation (BIM-focused). Several studies over the past decade and a half (including Guidera, 2007; Dupuis et al., 2008; Rassati et al., 2010; Yan et al., 2011; Becerik-Gerber et al. 2012; Irizarry et al., 2012; Mathews, 2013; Nawari et al., 2014; Charlesraj et al. 2015; Andrea Gutierrez-Bucheli et al. 2016; Zhang et al. 2018) have all reported similar approaches at different scales and we may consider BIM-aware and BIM-focused education to be firmly established in construction education.

Some examples of BIM-enabled education also exist. Wei and Wang (2018) reported BIM-enhanced safety training within a BIM-enabled environment for occupational awareness improvement. Hu (2019) transformed a traditional Building Materials and Construction Methods course using a BIM-enabled pedagogical approach and teaching platform to help students understand fundamental concepts. Witt and Kähkönen (2019) asserted that by enabling the use of real construction project data and simulating more realistic, multidisciplinary workflows in the educational environment, BIM-enabled education offers opportunities to enhance construction education didactics to better align graduate's competences with emerging and future needs, particularly those arising from the digitalization of the construction industry. However, in contrast to BIM-focused learning, the number of documented cases of BIM-enabled education is relatively few, suggesting that this stage of BIM education is still emerging.

This study therefore investigates the potential for applying BIM-enabled education, specifically, in the construction management curriculum in higher education institutions and to the topic of project cash flows.

Methodology

An action research design was selected as this offers a systematic procedure to address teaching improvements in their educational setting (Creswell 2012). Within BIM education, action research approaches have been applied effectively in contexts where there is a need to concurrently assist in implementation while interacting with staff and students to determine what works, what doesn't and to trial improvements (e.g., Williams et al. 2004; Puolitaival and Forsythe 2016). In this research, the five-phase process recommended by Susman and Evered (2016) was followed as listed below and represented in Figure 1:

Figure 1. The action research process applied.

1. Diagnosing – identifying or defining the problem under study;

2. Action planning – selecting and developing a research strategy to implement the intervention;

3. Action taking – identifying potential solutions to the problem and selecting and implementing the most appropriate of them.

4. Evaluating – analysing and assessing the results of the actions taken;

5. Specifying learning – recognising and documenting the outcomes of the research in order to understand what problems are solved, which objectives are met, and/or what theories are supported or require revision.

This study comprised two iteration cycles of the five-phase process and this enabled the educational intervention to be further refined. The first cycle was conducted in the Autumn semester of 2019. After reflection and identifying the need for further improvement, a second cycle was carried out with a different group of students but within the same course in the Autumn semester of 2020.

Educational intervention (case) description

The education intervention is a BIM-enabled Cash Flow Exercise which was implemented in 2 cycles of an action research process at Tallinn University of Technology within a Construction Investments course. The Construction Investments course objectives are:

1. to introduce students to investments, investment appraisal methods, and the bases for making investment decisions in the built environment;

2. to explore the current global, regional, and national contexts in which construction investments are being made;

3. to make students aware of the policies, programmes, and incentives/disincentives which influence investment in the built environment;

4. to give students an appreciation of the nature and motivations of investors and types of construction-related investment projects;

5. to raise students' awareness of the social and environmental aspects of built environment investments; and,

6. to familiarise students with project finance concepts and recent developments.

Students of this course are typically either fourth year students following an integrated, 5-year Master of Science in Civil Engineering degree programme or taking a 2-year master’s degree having completed their studies to bachelor's degree level previously, in a separate programme. Their industry and BIM experience varies from none at all to a high level of professional practice. In addition, as the course is one of few in the Structural Engineering and Construction Management study programme that is delivered in English at this university, it attracts international exchange students who may be from other disciplines.

The BIM-enabled Cash Flow Exercise is a learning experience in which students develop, analyse and optimise company cash flows in the context of a construction project scenario. A "5 D" (3 spatial dimensions + time + cost) BIM model serves as the learning object around which the learning activities take place. The exercise is arranged as group work where teams of participants assume different company roles (Developer, Main Contractor, Subcontractors) akin to those typical in the construction industry. The specific aim of the exercise is to collaboratively negotiate a global cash flow solution which enables all the companies to meet their (cash flow) objectives. This is achieved in 4 steps:

1. development of the project cash flows for each of the companies involved in construction;

2. analysing the developed cash flows and whether they meet specified (cash flow) objectives for each company (in terms of adequacy of return/profitability, credit limits, etc.);

3. negotiating a global solution which enables all the companies to meet their objectives and requirements; and, if this solution requires any changes to the (time scheduling) information in the 5D BIM model, then;

4. specifying the changes necessary to the BIM (see Figure 2 for context).

   Figure 2. BIM-enabled learning environment (BLE) framework for Cash flow analysis and optimisation adapted from Witt and Kähkönen (2019).

   

Table 1 summarises the attributes of both cycles of the intervention. The number of registered students for the courses in both cases exceeded the number of student participants in the exercise as exercise participation was not a direct pre-requisite for passing the course. In cycle 1, data collection was limited to a post-implementation questionnaire for students as well as recording the observations and reflections of staff. This was further developed in cycle 2 to include a pre-implementation questionnaire for students as well as development of the post-implementation questionnaire.

Table 1. Cycles attributes.

Cycle attribute	Cycle 1	Cycle 2
Semester	Autumn 2019	Autumn 2020
Number of groups	5	5
Number of participants in groups	G1 = 3, G2 = 3, G3 = 3, G4 = 2, G5 = 2	G1 = 6, G2 = 6, G3 = 6, G4 = 6, G5 = 6
Time allocated for the exercise	1 h and 30-min in one class session	3 h and 40 min in 3-class sessions: 2 of 1 h & 30 min each and an additional 40 min from a 3rd class session
Data collection	Post-implementation questionnaire, observation, and facilitators’ self-reflection.	Pre- and post-implementation questionnaire, observation, and facilitators’ self-reflection

Cycle 1

Cycle 1 of the action research began in late 2018 with the initial conception of the BIM-enabled Cash Flow Exercise which was subsequently implemented in the Autumn semester of 2019.

Diagnosing

Earlier research, including Witt and Kähkönen (2019), Olowa et al. (2021), had established the importance and potential benefits of BIM-enabled education and the cash flow topic appeared to offer a suitable and conveniently bounded opportunity to trial a BIM-enabled education intervention at the topical level. The previous method of cash flow teaching, as analysed through document analysis and lecturer observations/reflections, saw the application of cash flow calculations to a simple investment example which was relevant to a construction firm. In contrast, a BIM-enabled approach to cash flow teaching allowed the application of cash flow concepts to a scenario which was fully embedded within a construction project.

Action planning

Action planning involved the definition of learning outcomes, teaching and learning activities, feedback and assessment strategies for the intervention. The teaching and learning activities were premised on the principles of BIM-enabled learning as recommended by Underwood et al. (2013) where the focus is neither on informing students on the existence and the changes that BIM is bringing to the construction industry nor on how to use BIM as software applications but rather in deploying and leveraging BIM in classrooms to educate students of AEC-FM disciplines on construction management and engineering concepts to improve efficiency in both teaching and learning, increase students’ motivation and encourage lifelong learning. The learning activities and anticipated behavioural outcomes were defined with reference to Bloom’s taxonomy (Bloom 1956) and are shown in Figure 3.

Figure 3. Learning activities and outcomes for the BIM-enabled intervention.

Action taking

The intervention was implemented in the Autumn semester of 2019. The learning activity focused on collaborative learning and students worked in groups representing 5 different construction industry stakeholders, namely: developer, contractor, cast-in-place subcontractor, precast subcontractor and equipment subcontractor. A 5-D BIM model of a multistorey car park (Figure 4) was used as the learning object (Torrente et al. 2009). It was important for the model to have both time and cost data so that students could extract, process and then feed-back the processed data into the model. With reference to Figure 2 (above) Knowledge 1 and Knowledge 2 represent what the group members knew before and after the exercise, respectively. Data 1 represents all the data in the BIM model and in the additional instructional materials provided to students to enable them to undertake the exercise. Once their tasks were successfully complete and a global solution to the project cash flow had been found that satisfied all stakeholders' cash flow objectives, the resulting Data 2 could then be fed back into the BIM model. With that, the learning activity in the form of one iteration of the BIM workflow was completed.

Figure 4. 5-D BIM model of a multistorey car park.

Evaluating

The evaluation of this cycle of the study was based on a short questionnaire survey, facilitator’s reflection notes and outcome of the exercise. The questionnaire was an online “google form” with 6 questions requiring short text responses and the link was sent to all students who had participated in the exercise. From the 17 student participants, 6 questionnaires were returned. For the purpose of comparative analysis, the responses to the 5 closed questions have been quantified using the following subjective categories: 5 = Absolute Yes, 4 = Qualified Yes, 3 = Neutral, 2 = Qualified No and 1 = Absolute No. The summary of results is shown in Table 2.

Table 2. Implementation survey result of students’ satisfaction in Cycle 1 (N = 6).

Question	Mean	Std. deviation
Was the purpose of the exercise clear?	4.50	0.548
Did you find the exercise interesting?	4.33	1.211
Was it helpful to link the exercise to 'real' project data?	4.33	0.816
Was it helpful to link the exercise to the BIM workflow?	3.50	1.378
Did the exercise complement previous course materials or lessons?	4.33	0.816

The 6th question was an open question that asked participants how the exercise could be improved, and the comments received related mainly to increasing the time allocated for the exercise. For example, “There should be a longer class if it's possible”, “Reserve enough time so that everyone could have time to understand the whole process”.

Specifying learning

This section reports on some of the important lessons learnt in the first iteration of the intervention. Firstly, it was found that the exercise instructions were clearly understood by the participants although one participant did suggest that the exercise could have been introduced in an earlier class to allow for greater familiarity with the instructions before the actual exercise was carried out. Secondly, that the learning approach to the exercise was preferred by the participants compared to traditional classroom teaching. The participants further appreciated the linking of the exercise to the (near-to-real-life) 5-D BIM project model. Although the exercise demonstrated the BIM workflows related to project cash flow calculations, analysis, and negotiations, the correspondence of the exercise tasks with industry BIM work flows was not obvious to some participants.

The most important lesson drawn from Cycle 1 related to lesson plan improvement with respect to time allocation. The time allocated for the exercise was inadequate and this resulted in participants being unable to fully complete it. Only a few groups were able to complete their information input into the shared spreadsheet. This corresponds to completing step 2 of the 4-step exercise process as described above. This input into the shared spreadsheet by all groups was prerequisite to negotiating a global cash flow solution (step 3) and then updating the BIM model information (step 4). Thus, in its first cycle, the exercise ended with an almost complete shared spreadsheet. The lecturer did, however, demonstrate in the first minutes of the subsequent class how all stakeholders' cash flows could meet their requirements if the works were rescheduled and that this rescheduling would then need to be reflected in a revised 5 D BIM model.

Cycle 2

Diagnosing

The lessons learnt and challenges encountered in the first iteration provided useful insights into conceptualising, designing, and implementing topical level BIM-enabled education and they informed the next phase of action planning.

Action planning

The proposed learning experience and standard within the existing contextual factors were maintained with a few adjustments. Notable refinements were to the time allocated to the learning activities and to the data collection and evaluation arrangements.

Following from the experience of the first iteration, the overall activity time for the actual intervention was increased from 1-h and 30-min in a single class session to 3-h in 2-class sessions of 1-h and 30-min each. Having 2 class sessions greatly improved the likelihood of exercise completion and also introduced a 1-week time gap mid-exercise which allowed additional catching up time if the target half-way point had not been achieved within the first class. In addition, a desktop study was conducted to explore how best to comprehensively evaluate BIM-enabled education at a topical level.

The group size was left fluid for two major reasons: firstly, students were offered the option of virtual participation in this cycle due to COVID-19 which prevented some students from being physically present in the classroom. Secondly, student groups had been formed at the beginning of the semester and it was convenient to maintain these groups for the BIM-enabled Cash Flow Exercise, particularly as the groups contained physically present, virtually present and non-participating students and these could change at short notice. For these reasons, it was difficult for the facilitators to accurately determine the number of active student participants for each stakeholder group and led to larger group sizes than would be considered optimal.

Action taking

Cycle 2 was carried out in the Autumn of 2020 with a maximum participation of 30 students in 5 groups of 6 members each. As previously, in the lead up to the exercise, students were introduced to topics (e.g., cash flow development principles and investment appraisal methods such as net present value (NPV), internal rate of return (IRR), etc.) which were basic to them carrying out the cash flow exercise. Additional information and understanding required to carry out the exercise (stakeholders' financial status and project cash flow objectives, relevant conditions of contract relating to payment terms, retention and defects liability periods, etc.) were explicated by the facilitators at the start of the exercise.

Evaluating

The feedback and assessment strategies used followed the evaluation framework recommended in an earlier study (Olowa et al. 2021) for evaluating BIM for Construction Education (BfCE) interventions. The evaluation of this iteration was approached more methodically and systematically than that adopted for Cycle 1 and followed the processes of: (i) preparation, (ii) selection of evaluation tools, (iii) design and develop evaluation, (iv) implement evaluation and (v) review implementation (under both pre- and post-implementation). The strategy of designing and hosting the questionnaire survey as a Google form was maintained from Cycle 1, however, for Cycle 2, a pre-implementation questionnaire survey was included. While responding to the questionnaires was still not compulsory for participants, it was strongly encouraged and the number of returned questionnaires: 17 and 18 for pre- and post-implementation, respectively, represented an increase in response rate compared to Cycle 1. The lessons learnt in this cycle are based on the analysis of the responses to the questionnaires as shown in Table 3, in conjunction with class observations and facilitator’s reflections in the specifying learning section. (The same numerical scale of responses applies as per the responses to Cycle 1 questionnaire: 5 = Absolute Yes, 4 = Qualified Yes, 3 = Neutral, 2 = Qualified No and 1 = Absolute No.)

Table 3. Implementation survey results in Cycle 2.

Questions	Pre-implementation		Post-implementation
	Mean	Std. deviation	Mean	Std. deviation
Have you been taught about Building Information Modelling (BIM) in any previous courses?	2.83	2.007	NA	NA
If yes, how would you rate your current understanding of BIM?	1.64	1.391	NA	NA
Do you enjoy working in groups/teams?	4.56	0.856	4.59	1.004
Do you find it useful for your own understanding to discuss problems/calculations/solutions with your peers?	4.56	1.149	4.82	0.728
Do you understand how cash flow and cash flow calculations relate to construction projects?	2.94	1.097	3.92	0.493
Do you understand how cash flow relates to a BIM workflow?	1.69	0.910	3.33	0.924
Do you understand how different companies involved in a construction project can collaborate in order to optimize the project cash flow?	2.28	1.239	4.14	0.660
How important are cash flows to companies involved in construction projects?	NA	NA	4.56	0.482
Would you be interested if a similar BIM workflow is used to explain other subjects (concepts) apart from cash flow?	NA	NA	4.06	0.906
Was the purpose of the exercise clear?	NA	NA	4.28	1.074
Did you find the exercise interesting?	NA	NA	4.83	0.707
Did the exercise complement previous course materials or lessons?	NA	NA	4.94	0.243
Did the exercise improve your overall knowledge of cash flow?	NA	NA	4.56	0.856
Was it helpful to link the exercise to 'real' project data?	NA	NA	5.00	0.000

*NA = Not applicable.

Specifying learning

Important learning to be specified from Cycle 2 includes that, although this intervention was not designed to directly teach students about BIM models, it was still considered necessary to find out if the students had any previous knowledge of BIM. This questioning revealed that approximately equal numbers of participating students had prior BIM knowledge as didn't have any prior BIM knowledge. Notwithstanding the discipline of most of the student participants being civil engineering, this was not a very surprising result because almost half the students were Erasmus exchange students and many of them came from different disciplines but took the course because it was in English. By introducing some control questions in both the pre- and post-implementation questionnaires, the exercise also seemed to have shifted the students’ understanding of cash flow and how cashflow calculations relate to construction projects in a positive direction. The same observation in knowledge shift was made concerning the relationship between cash flow, BIM workflow and project cash flow optimisation. The exercise was perceived to be understood and considered as interesting by the students. The impact of the current COVID-19 pandemic was that the possibility of virtual participation was taken by some students whereas, in the first iteration, all students participated physically. This demonstrates that such an exercise could also be carried out (entirely) virtually, if necessary.

Results

Evaluation of the intervention was based on the participants' feedback from questionnaire surveys in conjunction with the facilitators’ observations and reflections. These indicate that the intervention accomplished its intended learning objectives and that the intervention itself and instructional strategies employed were favourably assessed by participants. Specifically, as shown in Tables 2 and 3, students' understanding was seen to improve as a result of the exercise.

Cycle 1 outcomes

This cycle was more or less to establish a proof of concept and, as such, little emphasis was placed on the evaluation of the exercise as a whole. Nevertheless, the facilitators asked the participants to express their opinions on the exercise’s clarity, interest generation, relation with reality, association with BIM workflow, and reinforcement of previous knowledge.

Was the purpose of the exercise clear?

The participants’ response to the above question gave a mean score of 4.50 when measured on a 5-point Likert scale using 5 = Absolute Yes, 4 = Qualified Yes, 3 = Neutral, 2 = Qualified No and 1 = Absolute No. This indicates that all the participants’ agreed that the individual group’s tasks, purpose and objectives in arriving at a globally negotiated and agreed project cash flow in the exercise were clear and understood.

Did you find the exercise interesting?

The participants’ average response to this question was 4.33. This means that most students agreed that the exercise was engaging.

Was it helpful to link the exercise to 'real' project data?

The ability to access and work with real project data is pivotal to this study and the majority of the participants, with an average score of 4.33, agreed that working through the exercise based on real project data was helpful.

Was it helpful to link the exercise to the BIM workflow?

This response had a lower mean score of 3.50 compared to the rest of the questions asked from the participants but still received an overall positive (yes) score. This may reflect the large number of participants who were not familiar with the BIM workflow and suggests a need for them to be educated in some BIM skills before the exercise.

Did the exercise complement previous course materials or lessons?

The researchers were curious to know if the exercise was helpful in reinforcing and integrating previously learned concepts - this being one of the goals of the intervention. The responses received show that majority of the participants, with mean score of 4.33, agreed that the exercise complemented previous course materials or lessons.

Cycle 2 outcomes

Do you understand how cash flow and cash flow calculations relate to construction projects?

Average rating before: 2.94; after: 3.92; statistically significant at p = 0.00. Based on the subjective categories earlier assigned i.e., 5 = Absolute Yes, 4 = Qualified Yes, 3 = Neutral, 2 = Qualified No and 1 = Absolute No; this infers that majority of the students were more or less neutral when first asked the question “Do you understand how cash flow and cash flow calculations relate to construction projects?” Whereas their response to the same question at the end of the exercise tended toward “Qualified Yes” which cannot be attributed to random chance after conducting statistical significance test with a p value of 0.00.

Do you understand how cash flow relates to a BIM workflow?

Average rating before: 1.69; after: 3.33; statistically significant at p = 0.00. Following similar classification of the responses into the five categories as in the previous question, the pre-implementation perception of the respondents tended toward “Qualified No” when asked the question “Do you understand how cash flow relates to a BIM workflow?” However, it tended toward “Qualified Yes” at post implementation with the difference in result not being due to random chance as confirmed by the statistical significance testing.

Do you understand how different companies involved in a construction project can collaborate in order to optimize the project cash flow?

When asked the question: “Do you understand how different companies involved in a construction project can collaborate in order to optimize the project cash flow?” Generally, the participants’ responses were toward a “Qualified No” before the intervention and toward an “Absolute Yes” after the intervention. Statistical test of significance showed that the observed change in knowledge was not due to random chance but rather because of the intervention.

It is also notable that the participant evaluations show an improvement from Cycle 1 to Cycle 2 (when Tables 2 and 3 values are compared) in terms of:

• students finding the exercise interesting (average rating Cycle 1: 4.33; Cycle 2: 4.83; not statistically significant, p =0.19).

• the helpfulness of linking the exercise to 'real' project data (average rating Cycle 1: 4.33; Cycle 2: 5.00; statistically significant at p=0.06).

• the exercise complementing previous course materials or lessons (average rating Cycle 1: 4.33; Cycle 2: 4.94; statistically significant at p=0.05).

Though one indicator did appear to drop slightly from Cycle 1 to Cycle 2:

• clarity of exercise purpose (average rating Cycle 1: 4.50; Cycle 2: 4.28; not statistically significant, p=0.26).

However, it should be taken into consideration that, for both cycles, participants rated all these indicators highly and, from the facilitators' reflections, Cycle 2 was definitely an improvement on Cycle 1 but there remains room for further improvement. Opportunities for improvement may be categorised into:

• Time considerations;

• Stakeholder group sizes and comparative workload issues;

• BIM model quality improvements;

• Learning environment.

Time allocation for the first iteration (1 ×90-min class) was clearly insufficient and this led to the exercise not being completed. The second iteration, in 2 ×90-min classes with a week's break between them, was a considerable improvement and allowed groups space to reflect and to catch up if they were behind after the first 90-min class but time was still insufficient and a further 40 min had to be added in order to achieve completion.

Discussion

BIM-enabled learning activities involve understanding project contextual characteristics through the BIM model and other associated project information (contractual arrangements, etc.) which are all incorporated into the activity (as inputs, boundary conditions, etc.) as these relate to the core of learning (Senaratne and Pasqual 2011). Students require time to explore and absorb this information and, particularly for pilot interventions at the topical level, such as the BIM-enabled Cash Flow Exercise, the relative amount of time needed for this is considerable. However, where numerous learning activities covering several topics can utilise the same project contextual information, the time needed will diminish proportionately and be relatively minor for course-level interventions and almost negligible if the approach were adopted for whole programmes. This later strategy is prevalent in BIM education as observed in the study of over 304 BIM education cases by Barison and Santos (2018) with the aim of providing students with clarity on conceptual issues through BIM visualisation thereby creating the advantage of not having to re-create BIM models for every curriculum in addition to fostering a valuable learning environment (Macdonald 2012; Lee and Hollar 2013).

A second set of time-related problems arose as a consequence of the stakeholder groupings and the difficulty in achieving a balanced workload. The concept of the grouping was intended to serve several pedagogical purposes. Firstly, to simulate the industry workflow by aligning students to specific industry roles and that working in groups would promote collaboration and active learning thus adding value to the exercise. Secondly, and in alignment with the observation by Hu (2019), instead of a step-by-step, piecemeal method that is characteristic of traditional learning styles, the new pedagogical strategy increased students' drive by presenting an all-inclusive and sophisticated view of the topic. Because of the steep learning curve and the availability of new technologies, the new generation of students are able to seek aid in a variety of ways on their own. This not only boosted learning confidence but also motivation (Hu 2019). The value of group working was largely confirmed by the students themselves in their positive responses to questioning about their group work (refer to Table 3). However, the group arrangements also led to time problems as some groups (the Developer and Main Contractor) had more complex calculations to perform than others (Subcontractors), some groups experienced calculation problems which took time to resolve and the mutual dependence for each other’s calculation outputs ultimately meant that some groups had to wait for others before they could finalize their calculations. This situation had been anticipated by facilitators in the preparation of both cycles and, to some extent, it was mitigated by selecting student groups perceived to be stronger to the more demanding roles and also providing more support to these groups during the exercise. However, more can and should be done to bring about workload balance between groups. For example, one alternative would be to have multiple projects in which each group undertakes a different stakeholder role thus enabling balance and a more even learning experience.

In addition, group sizes affected their performance, and, from the first iteration, it was understood by facilitators that the number of group members should ideally be limited to 2 or 3 where 3 is the optimal group size and 2 is preferable to 4. Establishing the optimal group size has always been a critical factor for consideration by project-based learning researchers. Davis and Miller (1996) recommended a group size of five, Henke (1985) group sizes of 3–5, while Barab et al. (2000) and Peterson and Myer (1995) report problems with group sizes of three and five. This suggests that group sizes of 4 or even number sizes are preferable (Helle et al. 2006). However, similar to the group size reported by Becerik-Gerber et al. (2012) and in contrast to the understanding gained from Cycle 1, there were up to 6 members in each group for the second iteration though the actual number of active participants in each group throughout the exercise was not possible to determine given the blended (face-to-face and virtual) nature of the groups under COVID-19 restrictions. The possibility of having two sets of 5 groups with a maximum of 3 members in each group could have been preferable, following the recommendations of Henke (1985), but such a strategy would have increased the risk of having a stakeholder group with 1 or no active members and that would have caused considerable disruption.

The 5-D BIM model was designed using Tekla Structures but, for the purpose of the exercise, only a free viewing tool was required and Tekla BIM sight (now Trimble Connect for Desktop) was used. Since the inauguration and subsequent development of Industry Foundation Classes (IFC) by the buildingSMART (former International Alliance for Interoperability—IAI) between 1995 and 2000 (Barison and Santos 2018), it should be noted that the .ifc file could also be opened with any other BIM software but this gave rise to interoperability issues in that data relating to cost and time were missing when the ifc file was opened in non-Tekla applications like Autodesk Revit and Navisworks Manage. This experience accentuates the argument of Zhang et al. (2013) that IFC is a complex and redundant data-modelling framework that requires precise implementation guidelines. The data from the BIM model needed to be 'cleaned up' after extraction from the industry foundation class (ifc) file format into the comma separated values (csv) file formats so that student participants could directly use it in spreadsheet applications with minimal (formatting and editing) effort especially since the intervention was time constrained. This underlines the need for high quality (error free, sufficiently and correctly detailed) BIM models in open, compatible formats that are fully readable with multiple software applications in order to support BIM-enabled learning activities.

Virtual instruction and participation were fundamental to this intervention, particularly in Cycle 2, and from the societal point of view, this is a trend that may well persist and should be further developed and promoted with time. Several authors have argued that the ability to use technologically supported simulation, and discussion forums are significant characteristics of e-learning environments (Fowler 2015; McGrath et al. 2018; Cai et al. 2019). Barari et al. (2020) re-affirm this position by stating that virtual teaching platforms with adequate facilities contribute to students’ collaborative teaching and peer learning. The present exercise employed multiple platforms in its delivery – for BIM model viewing and data extraction (Tekla BIM sight), for collaborative spreadsheet development (One Drive) and, in Cycle 2, virtual group work (MS Teams). This brought with it challenges in terms of accessibility and interoperability. Although virtual instructions have been reported by some studies to be beset with student frustration especially in BIM-focused learning (Becerik-Gerber et al. 2012), this was not the case in this study. As such, the authors opine that, for BIM-enabled learning to function efficiently, there is a need for a purpose-built, integrated BIM-enabled learning environment where all necessary capabilities are available in one place. Such a platform should enable model viewing, editing, storage, retrieval, etc. together with learning activity-specific operations (for the cash flow exercise, these primarily involve collaborative spreadsheet calculation capabilities).

Conclusion

Building Information Modeling (BIM) has significantly changed architectural, engineering, construction and facilities management (AEC-FM) practices and how the AEC-FM sector functions. Whereas, in industry practice, BIM has already become commonplace, how this important development might be applied effectively in higher education is still being determined.

This study sets out to explore how BIM can be leveraged to incorporate project-based experiential learning in construction management education and to evaluate students’ and facilitators’ perceptions of a BIM-enabled pedagogical intervention. A two-cycle, participatory action research study was carried out to pilot and refine an intervention at the topical level: the BIM-enabled Cash Flow Exercise. The responses received from student participant surveys and facilitators' reflections at the end of each cycle confirm the value of this type of BfCE approach. The topical level of the intervention has the advantage that there is no need for a drastic change in the existing curriculum, but the time needed for introducing the project contextual information is relatively high. Both the BIM-enabled learning activity design and the descriptive evaluation methodology used in this study will serve as a guide for researchers and future curriculum development of BfCE within AEC-FM education.

There are limitations relating to the generalizability of these results. The evaluation and conclusion on the success of this intervention are both premised on a short-term, cross-sectional study and a longitudinal study is required for more objective evaluation. Other contextual factors such as teacher–student and student–student relationships were not considered in this study. The possibility of introducing a control group exists and this was considered as an alternative to the use of a pre-implementation survey questionnaire which may have sensitised students to the purpose of the intervention and thus affected their (post-implementation) responses.

With the existing time constraints, the exercise scenario is very simple and the use of specific, real project data is quite limited. While making it easier to apply the recommended cash flow management process steps to it, this does reduce its correspondence to industry reality and thus it leaves a gap between the exercise and practical application in real industry projects.

To support this and other, similar BIM-enabled learning activities, particularly in distance learning and blended learning contexts, there is an emerging need for the specification and development of a BIM-enabled learning environment or platform in which the functionalities required to carry out BIM-enabled learning activities (e.g., BIM model viewing, data extraction, editing, etc. as well as activity-specific functions – collaborative spreadsheet applications for cash flow calculations) would be available. Future research, therefore, will focus on identifying these functional requirements and their corresponding technical requirements in order to derive a specification for an open, accessible and compatible BIM-enabled learning environment.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research was supported by the BIM-enabled Learning Environment for Digital Construction (BENEDICT) project (grant number: 2020-1-EE01-KA203-077993), the Integrating Education with Consumer Behaviour relevant to Energy Efficiency and Climate Change at the Universities of Russia, Sri Lanka, and Bangladesh (BECK) project (grant number: 598746-EPP-1-2018-1-LT-EPPKA2-CBHE-JP) and Strengthening University-Enterprise Collaboration for Resilient Communities in Asia (SECRA) project (grant number: 619022-EPP-1-2020-1-SE-EPPKA2-CBHE-JP), all co-funded by the Erasmus + Programme of the European Union. The European Commission support to produce this publication does not constitute an endorsement of the contents which reflect the views only of the authors, and the Commission cannot be held responsible for any use which may be made of the information contained therein.