Partnering with EMC Dell to Infuse Big Data Analysis into the Ministry of Public Administration of Republic of Slovenia
This case was submitted as part of the Call for Innovations, an annual partnership initiative between OPSI and the UAE Mohammed Bin Rashid Center for Government Innovation (MBRCGI)
The pilot project “Big Data Analysis for HR efficiency improvement” has been established as part of the strategy supporting data-driven public administration in Slovenia. It ran within the Ministry of Public Administration in collaboration with the company EMC Dell to learn how big data tool could improve the efficiency of HR data.
The pilot project has been established two common objectives: to increase the efficiency of public administration and to create a favourable environment for economic development. It was executed in collaboration with international company EMC Dell which in 2015 signed a letter of intent for cooperation with Ministry of Public Administration (MPA) of the Government of Republic Slovenia.
The Big Data pilot project launched in April 2016, with the objective to learn what a big data tool installed on the Governmental State Cloud Infrastructure could enable in terms of the research of HR data in the Ministry of Public Administration to improve our efficiency, develop organisational capacity, improve effectiveness and efficiency and staff satisfaction.
The project started with several introductory workshops aiming to select adequate data resources for further analysis as follows: data on employee’s time management (Codeks), ISPAP – salaries data, HR data and finance data (MFERAC) and data on public procurement. Since some databases contained personal data, the Ministry of Public Administration contacted the Slovenian Informational Commissioner. Analysis was performed within MPA premises on MPA IT infrastructure under strict security rules. Masking and anonymization of personal data was conducted to disable any possible identification on individual level. Furthermore, several interviews with data owners were conducted to clarify the content and to get interpretation of intermediate results. Additional external data sources -open data were added, such as historical weather data and geographical distinct. During prioritization process, three business initiatives (out of 40) were selected for further exploring. They were about HR management, real estate management and public procurement. One of them was the development of Employee Profiles. The employee’s profiles were segmented into relevant groups (clusters) based on multiple internal datasets containing employee job grade or specific role. HR features associated with high performance and low performance employees were analysed. Based on over 250 employee’s characteristics, five different groups (clusters) were designed containing different average performance scores. Also, open data - external data sources such as weather and geographic distances were included to provide additional insights on the behaviours of employees.
One of the fruits of this pilot project was that we learned that combining different data sources can enrich information and give us new value and new perspective. Using big data tools, we found new characteristics in our databases. Second learning point was that we should collect more data related to performance management systematically to enable more accurate analysis in this field in the future. Third learning point was that big data tools could enable us performing predictions for better planning process. Another learning point was related to personal data security where we gained knowledge how to handle personal databases in sense of data protection according to GDPR. The main lesson learned was about importance of establishing trust, firstly among IT experts and data sources owners which had to learn to listen each other and to understand each other’s needs. Secondly it was of utmost importance that our data source owners realized that they are the only one who can properly interpret their data and gained results since they have the most valuable knowledge on it.
What Makes Your Project Innovative?
Since some databases contained personal data, we had to consider GDPR and to design Impact assessment of privacy for Informational Commissioner to get approval for launching the project. In second step we had to mask and anonymise personal data properly. This was a challenge and it was done under strict security conditions with external partner in a close involvement of the employees - owners of data sources. Big challenge was the adequate interpretation and validation of mid-results and close collaboration of all involved parties: data owners, technicians (i.e., data scientists) and IT experts. Data owners were crucial part of the pilot working close with data scientists; this was challenge, but also a success factor of the project.
This pilot project confirmed that using big data tools could provide an effective and solid basis for predicting process, planning policies and decision-making process on all managerial levels in public administration based on existing internal data sources combined with open data. Gained experiences showed that big data analytics could help to improve the efficiency of the Ministry of Public Administration as basis for decision making by using different statistical and quantitative analysis.
What is the current status of your innovation?
Based on results and lessons learned from big data pilot, MPA realised that we should strengthen awareness on digital knowledge and to develop digital skills. Therefore, the first conference on Big Data Tools was performed in spring 2017 and since autumn 2017 also, new training program has been running by Ministry of Public Administration Administrative Academy in collaboration with experts University of Ljubljana and University of Maribor. The training program consists from different modules adopted for different focus groups such as managers, analysis and IT experts with different levels of complexity and as follows:
- Data science for beginners,
- Business Intelligence,
- Introduction to text mining,
- From questionnaires to content for decision-making,
- Protection of personal data and open data.
The public employees are very interested in the program, in six months (until now) our Administration Academy has performed over 15 workshops and over 150 participants have been trained in the field of data analytics.
Collaborations & Partnerships
In the Ministry of Public Administration, the project group consisted of 23 people - governmental officials were involved deriving from different parts of MPA: Secretariat, HR services, Financial services, Directorate for Public Sector, Public Procurement Directorate and IT Directorate. There were different profiles of the people on different levels collaborating in the project, both managers and professionals. From EMC Dell 5 experts participated.
Users, Stakeholders & Beneficiaries
Both parties in the project – MPA and EMC Dell participated in a project on their own expenses. From EMC Dell 5 experts participated. The EMC provided their knowledge on data science, MPA provided data, knowledge about data interpretation and both parties had win-win situation which enabled learning and gaining experiences. For MPA experiences with emerging technology and its benefits, for EMC Dell experience to work with governmental administration in the field of governmental HR, real estate management and public procurement.
Results, Outcomes & Impacts
This partnership changed the way the Ministry approaches public governance, as we gained knowledge what tools big data could bring. We shifted towards using more complex, sophisticated tools as our modus operandi. This pilot confirmed using big data tools could provide benefits supporting prediction, planning and decision-making process. Based on gained experiences we decided to launch project for implementation of business intelligence and data warehouse to support data based decision making for repetitive reports. To raise awareness training program by MPA Administrative Academy in collaboration with experts University of Ljubljana and University of Maribor was launched. Beside we are developing big data infrastructure as a sand box to enable testing in this field. We believe personal data security in terms of GDPR is one of the fundamental rights. On the other hand, using big data tools brings other undefined issues related to personal data security. It seems to be a gap in-between which we believe will be resolved in the next few years in sense of development and support of usage emerging technologies in public administration where we deal with great amount of personal data.
Challenges and Failures
Since we used emerging technology - big data tools, in the beginning of the project some of data sources were sceptic about the project. They meant that this should be done by informatics. By the time when they were strongly involved in the interpretation and validation of their own data, they changed their minds, they saw their crucial role and were very satisfied with the outcome of the project. Data owners were crucial part of the pilot working close with data scientists and interpreting gained results; this was challenge, but also a success factor of the project.
Conditions for Success
There are multiple conditions needed for success of such kind of project. Firstly, strong leadership and promotion of the main goal of the project from the highest level in organisation. Secondly, developing awareness and communication with employees is of outmost importance. Thirdly, close collaboration of all involved parties is very important. Our project group consisted from people from different departments of MPA such as: Directorate for Public Sector, Directorate for Informatics, HR, Finance, Law and Directorate for Public Procurement. On the other hand, 5 experts from EMC Dell covered also different roles such as data scientists, data presenter, interpreter etc.
Based on experiences of big data pilot we realized that using big data tools could be successful used for research and providing new ideas. On the other hand our public administration needs tools which enable support for repetitive tasks which are used for decision making. Therefore, we launched project for business intelligence and data warehouse which is planned to be horizontal service for all governmental agencies by 2023.
We continue to develop big data infrastructure as a sandbox for further testing activities on our governmental cloud. We share our knowledge and experiences within Slovenian governmental agencies and wider on several international events.
Furthermore, we will continue with training programme on data analytics within MPA Administrative Academy in collaboration with University of Ljubljana and Maribor.
Data security was very important part of this pilot project, especially from the point of view related to personal data and GDPR. We followed our national legislation in the field of personal data security and directions which we got from our Information Commissioner on anonymization and data masking.
In the phase of data preparation and processing is very important factor development of trust among employees: IT experts and “non-IT people” – internal users which have to understand each other, develop common language and learn to listen each other to be able to work together for the common goal. Therefore, also good communication, “speaking the same language” among employees within (and outside) the project is of outmost importance. We realized that close collaboration of all involved parties is very important for success of this kind of activities. Our project group consisted from people from different departments of MPA, people who usually do not work together, therefore establishing trust and open atmosphere among them was of crucial importance.
At beginning of the project some data source owners did not see their role in the project, they thought IT people should deal with the data without their participation. They were not aware of their knowledge of the content of the data which is an important part of this process. In later phases of the project when they were actively included in data preparation and interpretation they realized the benefits of this tools on one side and the value of their knowledge on the other side.