
One of the key challenges that have always faced the capital construction industry is the talent drain. Many of the experienced project management team are either retiring or changing their career path. The Covid-19 pandemic has accelerated this trend specially in countries that depends on migrant resources who have either found employment back at their home countries or have started their own businesses. This means that the massive knowledge captured by those professionals over their many years of experience in delivering capital construction projects will be wasted. In other words, Junior project team members will not have access to this priceless knowledge and they have to go through the same lengthy doing mistakes and learning process of their predecessors.
Machine learning can be used to create value from this knowledge and make it available for the project team members to better predict future outcomes from today results. For example, one of the most critical requirements of capital construction projects is to ensure that the quality of the inspected, verified and approved work in place is accurate. Not only this is required to approve interim payment certificates but it is also required to ensure the quality, safety, reliability and usability of the constructed building systems. The knowledge associated with the quality of work in place is captured in the business processes for technical submittals, non-conformance reports, work inspection requests and defects reported during the defect liability period. Of course, there are other business processes that also contributes to the knowledge associated with work in place.
The chart below shows the quality fulfilment process that is common to capital construction projects. It starts by ensure that all technical submittal requirements are identified followed by the technical submittal for shop drawings, material samples, installation method statement, etc. The NCR is the business process used to identify actions that do not comply to what had been specified in the project and submitted for approval. The WIR is the process that will be used to verify that was done is in accordance to what has been specified for the project. This will entitle the contractor to get paid for the work that had been inspected, verified and approved. The last business process which is used to capture defects for work that was approved and paid for. Those defects need to be analyzed as the organization might find it is necessary to update the WIR templates and even the project’s technical specifications to ensure that the causes of those defects are captured when the work in place was inspected.
To improve the useability of the data required for the machine learning training data, the captured data needs to be classified with the specification section that they are associated with. For example, the Construction Specification Institute (CSI) MasterFormat® provides a detailed technical specification structure that can be used to have a comprehensive classification system for work inspections, technical submittals, non-conformance reports and defects reports. In other words, all Technical Submittals, WIR, NCR and work defects templates should have a field for CSI specification.
In addition, those templates should also have fields to capture other data variables that could have an impact for what is being reported. For example, the Work Inspection Request (WIR) template should include details of the contractor who have executed the works and the supervision consultant who have accepted the completed works. In addition, they need to include details of the weather condition on the date and time when the inspection was done, location of the inspected work, criticality of the project schedule activity representing the inspected work, observations or remarks made by those who were involved in the inspection.
Another requirement for improved machine learning is eliminating the bias and variance in captured data. Having a predefined checklists for all work to be inspected, issued non-conformance reports and reported defects will not only ensure consistency in reported data but also to ensure that the data in unbiased. Those checklists need to be aligned with project’s technical specifications requirement as well as the contract subclauses that necessitates the issuance of a non-conformance report or reporting a defect. Eliminating the actions that could lead to bias and variance in captured data will improve the quality of predicted outcomes.
A Project Management Information System (PMIS) like PMWeb can be used to create those templates with the required data fields. For example, for the Work Inspection Request (WIR) for “09 20 13 Ceramic Tiling”, PMWeb custom form builder will be used to create a template to capture all needed data. For example, the header will include the specification system, location, Work Breakdown Structure (WBS) level, schedule activity associated for the work to be inspected, date and time of the inspection, contractor or subcontractor who is responsible for the work in place and other fields that could be needed.
The WIR template will also include tables to capture the data associated with the material received, pre-installation requirements and site preparation. For each category there will be a predefined list of items that need to be inspected along with the document that provides the content or details of what to be inspected like for example Method Statement, Shop Drawing, etc. For each inspection item, the inspector needs to advise of what was inspected was compliant or not, when it was re-inspected after rectifying remarks and comments made on the inspection.
Another table will be created for capturing the details of the inspection done for the completed work in place. This can be also classified into the type of completed work, for example if the ceramic tiling to done for walls or floors. For each inspection item, the inspector needs to advise of what was inspected was compliant or not, when it was re-inspected after rectifying remarks and comments made on the inspection.
For the templates of technical submittals, non-compliance report and reported defects, the checklist for each template will be in a different format than what used for the Work Inspection Request (WIR) template as the details of what needed to be inspected and verified will differ. For example, for a non-compliance report, the template header will include the fields for the specification system, location, Work Breakdown Structure (WBS) level, schedule activity associated for the non-compliant work, date and time of issuing the NCR, contractor or subcontractor who is responsible for the work for which an NCR was issued and other fields that could be needed. In addition, it will include the fields for What is the main reason for the NCR or what went wrong, Why the work doesn’t meet specifications, what can be done to prevent the problem from happening again, Explanation of corrective action taken or to be taken and the NCR Corrective Action Detail.
Although much of those data fields are usually common to NCRs, nevertheless, what makes this NCR different is the checklist of the Reasons that have led for Issuing the NCR. By having those reasons predefined, the risk of having biased and/or non-consistent data can be eliminated. Of course, the list of those reasons as well as the content of the other checklists created for technical submittals, work inspection requests and reported defected will be developed by the experienced project team members who have the knowledge in performing those activities. Those checklists must be also aligned with the technical specification documents, bill of quantities and subclauses of the contract agreement document.
The data needed to improve the training data of the machine learning could also come from other the data of other business processes. For example, there is the data that relates to project type, size, location, delivery method, entities involved, budget, duration, built up area, number of floors and others captured in the project module. In addition, there is the data from daily reports which will help to identify weather conditions, safety incidents, if there was stacking of trades, events that have disrupted work or have a direct or indirect impact on the work to be inspected. To improve the reporting of the reported data in all these templates including the daily report template, the data needs to be classified with the specification section it relates and location of the works.
The transactions associated with the Technical Submittal, WIR, NCS, Reported Defects, Daily Reports and other business processes might have supportive documents that need to be associated with their relevant transaction. The attachment tab for each business process will be used to attach all those supportive documents.
It is also highly recommended to add comments to each attached document to provide better understanding of what was the document for. The attachment tab allows the user to also link other records for business processes implemented in PMWeb as well as associate URL hyperlinks with websites or documents that are not stored in PMWeb document management repository.
To enforce accountability for the data captured in the many transactions for each of the above detailed business process, PMWeb workflow module will be used to create a workflow to formalize the review and approval tasks of those business processes. The workflow will map the sequence of the review and approval tasks along with the role or user assigned to the task, duration allotted for the tasks, rules for returning or resubmitting a document and available for each task. In addition, the workflow could be designed to include conditions to enforce the authority approval levels as defined in the Delegation of Authority (DoA) matrix.
For the machine learning training data to be of value, it needs massive volume of data to be captured. This requires making those business processes available for all projects that are managed by the organization. A platform like PMWeb enables capturing the complete on-going and completed projects’ portfolio that an organization could have.
About the Author
Bassam Samman, PMP, PSP, EVP, GPM is a Senior Project Management Consultant with 40-year service record providing project management, project controls services and project management information system to over than 200 projects with a total value in excess of US $100 Billion. Those projects included Commercial, Residential, Education and Healthcare Buildings and Infrastructure, Entertainment, Hospitality and shopping malls, Oil and Gas Plants and Refineries, Telecommunication and Information Technology projects. He is thoroughly experienced in complete project management including project management control systems, computerized project control software, claims analysis/prevention, risk analysis/management (contingency planning), design, supervision, training and business development.
Bassam is a frequent speaker in topics relating to Project Management, Strategic Project Management and Project Management Personal Skills. Over the past 40 years he has lectured at more than 350 events and courses at different locations in the Middle East, North Africa, Europe and South America. He has written more than 500 articles on project management and project management information systems that were featured in international and regional magazines and newspapers. He is a co-founder of the Project Management Institute- Arabian Gulf Chapter (PMI-AGC) and has served on its board of directors for more than 6 years. He is a certified Project Management Professional (PMP) from the Project Management Institute (PMI), a certified Planning and Scheduling Professional (PSP) and Earned Value Professional (EVP) from the Association for the Advancement of Cost Engineering (AACE) and Green Project Management (GPM).
Bassam holds a Masters in Engineering Administration (Construction Management) with Faculty Commendation, George Washington University, Washington, D.C., USA, Bachelor in Civil Engineering – Kuwait University, Kuwait and has attended many executive management programs at Harvard Business School, Boston, USA and London Business School, London, UK.