Loading…
Learning from post project reviews
Post Project Reviews (PPRs) can provide a valuable source of learning for project teams. They are also known by other terminologies such as project closeout, project post mortems, etc, and attempt to document the project experience – both good and bad. In order to reflect their importance, many cons...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Default Conference proceeding |
| Published: |
2010
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/2134/11302 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1818174269761257472 |
|---|---|
| author | Patricia Carrillo Jennifer Harding Alok Choudhary Paul Oluikpe |
| author_facet | Patricia Carrillo Jennifer Harding Alok Choudhary Paul Oluikpe |
| author_sort | Patricia Carrillo (1248684) |
| collection | Figshare |
| description | Post Project Reviews (PPRs) can provide a valuable source of learning for project teams. They are also known by other terminologies such as project closeout, project post mortems, etc, and attempt to document the project experience – both good and bad. In order to reflect their importance, many construction organisations now have policies towards the conduct of PPRs. The reports resulting from these PPRs are done with the best intentions of providing a rich and valuable source of learning. However, because many companies do not have the resources to examine their review reports, either individually or collectively, important insights are missed thereby leading to a missed opportunity to learn from previous projects. Text mining offers a potential solution to companies that do not have the resources to analyse these reports. Text mining analyses large volumes of text to identify patterns and trends in order to extract information and knowledge that could improve process, and identify both good and bad practice. Text mining is a development of knowledge discovery and data mining; the latter uses numerical data and has been used successfully in a range of industry sectors such as banking, manufacturing and retail to improve customer satisfaction. Text mining is a relatively new approach and uses unstructured text, as found in PPR reports. It is thus ideally suited to overcoming the problem with organisations possessing a large number of PPRs that may provide very useful information and knowledge without the requirement for extra human resources to analyse them. This paper investigates the potential use of text mining to identify vital sources of knowledge that can lead to learning from Post Project Reviews. Two UK construction contractors provided PPRs reports. The companies adopted radically different approaches to the style and content of their PPRs reports and thus provided an opportunity to investigate the success of text mining for different scenarios. In total 48 PPR reports were analysed. The companies’ reports were first pre-processed to allow then to be used in a text mining tool. The text mining tool also had to be customised, using ontologies, to suit the context of the reports. In addition, both companies were asked to identify key knowledge areas that are important to their businesses; these formed the basis of the key words and phrases that were used for text mining. Two techniques, namely Link Analysis and Dimensional Matrix Analysis were used to identify correlations between key words and phrases that appear across a range of different Post Project Review reports. The initial results are very promising because they help to identify links and trends that would otherwise be difficult to identify without a substantial amount of manpower. One of the advantages is the graphical representation of the strength of correlations between key words that makes it easy to select areas for further investigation. |
| format | Default Conference proceeding |
| id | rr-article-9428684 |
| institution | Loughborough University |
| publishDate | 2010 |
| record_format | Figshare |
| spelling | rr-article-94286842010-01-01T00:00:00Z Learning from post project reviews Patricia Carrillo (1248684) Jennifer Harding (1258389) Alok Choudhary (1251471) Paul Oluikpe (7175675) Mechanical engineering not elsewhere classified Knowledge management Learning Project reviews Text mining Mechanical Engineering not elsewhere classified Post Project Reviews (PPRs) can provide a valuable source of learning for project teams. They are also known by other terminologies such as project closeout, project post mortems, etc, and attempt to document the project experience – both good and bad. In order to reflect their importance, many construction organisations now have policies towards the conduct of PPRs. The reports resulting from these PPRs are done with the best intentions of providing a rich and valuable source of learning. However, because many companies do not have the resources to examine their review reports, either individually or collectively, important insights are missed thereby leading to a missed opportunity to learn from previous projects. Text mining offers a potential solution to companies that do not have the resources to analyse these reports. Text mining analyses large volumes of text to identify patterns and trends in order to extract information and knowledge that could improve process, and identify both good and bad practice. Text mining is a development of knowledge discovery and data mining; the latter uses numerical data and has been used successfully in a range of industry sectors such as banking, manufacturing and retail to improve customer satisfaction. Text mining is a relatively new approach and uses unstructured text, as found in PPR reports. It is thus ideally suited to overcoming the problem with organisations possessing a large number of PPRs that may provide very useful information and knowledge without the requirement for extra human resources to analyse them. This paper investigates the potential use of text mining to identify vital sources of knowledge that can lead to learning from Post Project Reviews. Two UK construction contractors provided PPRs reports. The companies adopted radically different approaches to the style and content of their PPRs reports and thus provided an opportunity to investigate the success of text mining for different scenarios. In total 48 PPR reports were analysed. The companies’ reports were first pre-processed to allow then to be used in a text mining tool. The text mining tool also had to be customised, using ontologies, to suit the context of the reports. In addition, both companies were asked to identify key knowledge areas that are important to their businesses; these formed the basis of the key words and phrases that were used for text mining. Two techniques, namely Link Analysis and Dimensional Matrix Analysis were used to identify correlations between key words and phrases that appear across a range of different Post Project Review reports. The initial results are very promising because they help to identify links and trends that would otherwise be difficult to identify without a substantial amount of manpower. One of the advantages is the graphical representation of the strength of correlations between key words that makes it easy to select areas for further investigation. 2010-01-01T00:00:00Z Text Conference contribution 2134/11302 https://figshare.com/articles/conference_contribution/Learning_from_post_project_reviews/9428684 CC BY-NC-ND 4.0 |
| spellingShingle | Mechanical engineering not elsewhere classified Knowledge management Learning Project reviews Text mining Mechanical Engineering not elsewhere classified Patricia Carrillo Jennifer Harding Alok Choudhary Paul Oluikpe Learning from post project reviews |
| title | Learning from post project reviews |
| title_full | Learning from post project reviews |
| title_fullStr | Learning from post project reviews |
| title_full_unstemmed | Learning from post project reviews |
| title_short | Learning from post project reviews |
| title_sort | learning from post project reviews |
| topic | Mechanical engineering not elsewhere classified Knowledge management Learning Project reviews Text mining Mechanical Engineering not elsewhere classified |
| url | https://hdl.handle.net/2134/11302 |