Periodic Reporting for period 1 - IT-TRAS (INTEGRATED TIME TRACKING SYSTEM)
Reporting period: 2019-12-09 to 2021-06-08
Time tracking is an important component of efficient management, for assessing the efficiency of processes and tools, as well as individuals. Measurement of time spent on specific tasks is particularly difficult in the third and fourth sector of the economy and is further complicated by the extensive use of computers which facilitates the constant switching back and forth between tasks, rather than closing one before moving to the next.
Whether the goal is accurate invoicing or measuring the efficiency of human and other resources, the standard way of tracking time spent on various work tasks is through Global Timesheets, in other words a form of self reporting. However, the task of manually completing timesheets is time-consuming in itself and difficult to achieve accurately. This inaccuracy in manually reported timesheets has been calculated to cost millions per year. Furthermore, timesheet completion has been found to be a significant cause of work-related anxiety.
The problem of manual timesheets has been the focus of various software solutions aiming to provide automated digital time tracking. However, most of these solutions still depend on frequent user intervention to indicate switching from one task to another etc. This defines the overall objective of the IT-TRAS project: minimization of manual intervention in time-tracking through a) the gathering of data from different sources in order to work seamlessly in different locations and platforms; b) automatic adjustment to different tasks/projects; c) automatic time distribution.
Automated and accurate time-tracking is expected to significantly reduce time spent on the task of reporting, minimize inaccuracies and therefore financial losses and lessen anxiety and frustration stemming from these problems.
Whether the goal is accurate invoicing or measuring the efficiency of human and other resources, the standard way of tracking time spent on various work tasks is through Global Timesheets, in other words a form of self reporting. However, the task of manually completing timesheets is time-consuming in itself and difficult to achieve accurately. This inaccuracy in manually reported timesheets has been calculated to cost millions per year. Furthermore, timesheet completion has been found to be a significant cause of work-related anxiety.
The problem of manual timesheets has been the focus of various software solutions aiming to provide automated digital time tracking. However, most of these solutions still depend on frequent user intervention to indicate switching from one task to another etc. This defines the overall objective of the IT-TRAS project: minimization of manual intervention in time-tracking through a) the gathering of data from different sources in order to work seamlessly in different locations and platforms; b) automatic adjustment to different tasks/projects; c) automatic time distribution.
Automated and accurate time-tracking is expected to significantly reduce time spent on the task of reporting, minimize inaccuracies and therefore financial losses and lessen anxiety and frustration stemming from these problems.
M1 started with the relocation of the Innovation Associate and his integration in the Rhodes Centre for History and Social Research SCE and its research team. His hiring was registered with the competent authorities and he was registered with a GP. The kick-off meeting took place on 27.12.2019. The review of the state of the art also started during this month.
M2 saw the drafting of the tailored training plan, which was adjusted to suit the Innovation Associate’s background and the expected needs of the project. Following some market research a custom course was designed in collaboration with K2 lifelong learning campus. The state of the art report was completed, providing guidance for the remainder of the project.
M3 was dedicated to analysis of user requirements and usage scenarios.
M4 brought unexpected obstacles due to the COVID-19 pandemic and the subsequent restrictions. Work at the Research Centre had to continue online, with activities such as fieldwork and interviews suspended. Tasks pertaining to IT-TRAS, such as shadowing and data collection had to be adjusted accordingly.
In M5 work was done concerning URL keyword identification and filtering.
In M6 the team started working on the clustering algorithm. At the same time the Innovation Associate started attending the core training.
M7: work continued on algorithm development, analysis of requirements through remote shadowing and testing of algorithm progress. Core training continued.
M8 brought a major upheaval as the Innovation Associated resigned from the project. The project was suspended, while at the same time a new hiring process was initiated..
M9: Dr Despoina Panoglou, was hired as the new Innovation Associate and registered with the competent authorities. She immediately started work on understanding the code prepared by the previous Innovation Associate. At the same time the tailored training plan was redesigned taking into account the new Innovation Associates background and the project’s needs as determined by the development achieved up to that point. Dr. Panolgou started attending the training shortly afterwards.
M10: The kick-off meeting was held on 05.03.2021. The Innovation Associate started work on optimization of the URL and keyword analysis.At the same time she started attending the core training programme.
M11: the IA delivered a new clustering algorithm and connected it with time identification.
M12 was dedicated to packing the software into an executable file; testing for functionality under various circumstances; and fine tuning of the clustering algorithm based on test results.. At the same time the Innovation Associate successfully completed the tailored training.
M2 saw the drafting of the tailored training plan, which was adjusted to suit the Innovation Associate’s background and the expected needs of the project. Following some market research a custom course was designed in collaboration with K2 lifelong learning campus. The state of the art report was completed, providing guidance for the remainder of the project.
M3 was dedicated to analysis of user requirements and usage scenarios.
M4 brought unexpected obstacles due to the COVID-19 pandemic and the subsequent restrictions. Work at the Research Centre had to continue online, with activities such as fieldwork and interviews suspended. Tasks pertaining to IT-TRAS, such as shadowing and data collection had to be adjusted accordingly.
In M5 work was done concerning URL keyword identification and filtering.
In M6 the team started working on the clustering algorithm. At the same time the Innovation Associate started attending the core training.
M7: work continued on algorithm development, analysis of requirements through remote shadowing and testing of algorithm progress. Core training continued.
M8 brought a major upheaval as the Innovation Associated resigned from the project. The project was suspended, while at the same time a new hiring process was initiated..
M9: Dr Despoina Panoglou, was hired as the new Innovation Associate and registered with the competent authorities. She immediately started work on understanding the code prepared by the previous Innovation Associate. At the same time the tailored training plan was redesigned taking into account the new Innovation Associates background and the project’s needs as determined by the development achieved up to that point. Dr. Panolgou started attending the training shortly afterwards.
M10: The kick-off meeting was held on 05.03.2021. The Innovation Associate started work on optimization of the URL and keyword analysis.At the same time she started attending the core training programme.
M11: the IA delivered a new clustering algorithm and connected it with time identification.
M12 was dedicated to packing the software into an executable file; testing for functionality under various circumstances; and fine tuning of the clustering algorithm based on test results.. At the same time the Innovation Associate successfully completed the tailored training.
The overview of the state of the art performed by the Innovation Associate can be summarized as follows:
- Most available systems offer little or no true automation.
- There is minimal use of ML or AI in the time-tracking field.
- The one system that does use ML is severely limited in terms of identifying different tasks and taking into consideration work performed on different devices (laptop, desktop, tablet etc)
The IT-TRAS project has made progress in automation, particularly in terms of identifying and separating different tasks and assigning work time. It has also made progress in terms of keeping track of work on the same task on different devices, through the possibility of combining datasets and analyzing them together.
IT-TRAS has produced a software program that is capable of recognizing different work tasks through the use of clustering algorithms running on keywords extracted from the original datasets. It is based on the information stored in browser history files.
The impact of this progress lies with the possibility of more accurate, less time-consuming time-tracking, particularly in the third and fourth sectors of the economy. This in turn would have a great impact on efficiency measurement, of which time-tracking is a key component, providing managers with a powerful tool, but also freeing up time for employees and freelancers. Overall the expected impact may be summarized as: less time wasted; fewer mistakes; better time -tracking and management.
- Most available systems offer little or no true automation.
- There is minimal use of ML or AI in the time-tracking field.
- The one system that does use ML is severely limited in terms of identifying different tasks and taking into consideration work performed on different devices (laptop, desktop, tablet etc)
The IT-TRAS project has made progress in automation, particularly in terms of identifying and separating different tasks and assigning work time. It has also made progress in terms of keeping track of work on the same task on different devices, through the possibility of combining datasets and analyzing them together.
IT-TRAS has produced a software program that is capable of recognizing different work tasks through the use of clustering algorithms running on keywords extracted from the original datasets. It is based on the information stored in browser history files.
The impact of this progress lies with the possibility of more accurate, less time-consuming time-tracking, particularly in the third and fourth sectors of the economy. This in turn would have a great impact on efficiency measurement, of which time-tracking is a key component, providing managers with a powerful tool, but also freeing up time for employees and freelancers. Overall the expected impact may be summarized as: less time wasted; fewer mistakes; better time -tracking and management.