Expert opinions, INVESTMENT CLIMATE, TECHNOLOGY

Data driven: Hidden trump cars contributing to the efficiency of public administration

Russia is gearing up to launch a new national project, the Data Economy, aimed at overhauling the operational framework of regional administrations and instilling a culture of data-driven decision-making (Data-Driven Public Sector). This ambitious project necessitates the adoption of new IT tools by regions, the enhancement of staff capabilities to utilize them effectively, and, crucially, a paradigm shift in data collection processes and approaches through the cultivation of a new data management culture.

Pavel Bednyakov / RIA Novosti

Data handling culture

The Data Economy national project advocates for the shift from merely accumulating data to actively applying them in practice. Vast quantities of data are gathered online in operations centers of regional heads and ministries. Instead of merely showcasing them on BI system dashboards based on 20 key performance indicators for senior officials, these data should evolve into a pivotal strategic asset and a source of economic value for the population and backbone enterprises.

As a result, this national project acts as a catalyst for enhancing existing BI solutions by integrating new Decision Support Systems (DSS). These advanced DSS incorporate features for building master data repositories capable of modeling and resolving analytical issues, managing and detecting both explicit and implicit data dependencies, and providing comprehensive decision-making recommendations.

Next, we will illustrate how any Russian region can swiftly and confidently establish data-centric management using four steps and decision support systems. This approach will facilitate the development of new electronic services and public administration processes centered around data, offering comprehensive decision-making recommendations.

Step 1. Creating centralized data storage

Initially, creating a centralized data storage at the operations center might appear laborious and resource intensive. However, it’s a fundamental requirement without which fostering a Data-Driven culture and leveraging its benefits becomes impossible.

Regardless of where the Decision Support System is being deployed, the cornerstone always remains the reliability and quality of data, a facet often lacking in contemporary operations centers of regional ministries and heads. This deficiency stems from a historical trend over the last three decades, where information systems were constructed around business processes, resulting in redundant data requests and reporting. Many departments persist in gathering and processing identical information for various report forms, leading to discrepancies in the outcomes depending on the data sources and analysis parameters employed. For instance, statistics on birth rates gathered from maternity hospitals, accounting for the number of newborns, may diverge from the data consolidated in civil registries, tallying the count of infants issued birth certificates. While both sources are credible, disparities may arise if some expectant mothers opt for maternity hospitals in different regions, prompting the child’s registration in their respective birthplace.

There is indeed a solution to the challenge posed by redundant data and metrics. It entails leveraging the DAMA DMBOOK framework, understanding how to craft context diagrams delineating each knowledge area, detailing data sources, methodologies, tools, and metrics, as well as clarifying data consumers and modification protocols, with a crucial focus on accurate interpretation. Subsequently, this knowledge is digitalized and transferred to a dedicated Data Lake repository, where regional operations center data is described layer by layer. This project encompasses several thousand metrics sourced from regional and municipal systems, necessitating an average of up to 6 months of meticulous effort.

Step 2. Scenario modeling

After accumulating data and removing duplicates, we can now proceed to their actual application through scenario modeling.

Implementation of decision support systems is nothing new. However, currently only federal executive authorities can boast successful cases of building and using DSS that assist decision-makers in answering questions like: What will happen? What happens in case…? What should be done to…? What are the ways of optimal resource allocation? and others.

National Crisis Management Center of the Russian Emergencies Ministry was among the first to implement a DSS, which has successfully operated for over a decade. DSS have also been successfully implemented at the regional level – in Moscow, the Republic of Tatarstan and the Kaliningrad region, although operating with a limited number of scenarios.

The limitations of scenario modeling are due either to functional capabilities of the Business Intelligence (BI) Platforms or the data sources. Advanced DSS products allow for modeling and solving any analysis tasks, as well as managing and spotting explicit and implicit data dependencies through certain built-in tools designed to seek dependencies, identify influencing factors, select optimal parameters, carry out classification and ranking, detecting anomalies, and make forecasts.

Step 3. Developing a data-driven mindset

Starting this year, Russia will introduce a data-centric management approach, with new electronic services and governance processes to become data-based. In advanced countries, digital governments are already operating to build client-centric services. This requires public employees to eliminate bureaucracy, ensure horizontal interdepartmental relations and prompt communication between regional organizations, institutions and private businesses, as well as to create open datasets. For instance, Singapore’s government contends for services to be fully digitalized at all levels, constantly experimenting with flexible technologies and testing services for citizens and businesses with the use of sandboxes.

Although very scarce, there are successful customer centricity cases in Russia as well. The implementation of DSS as part of public administration’s digital transformation will contribute to building a data-centric government in each Russian region within the next three years.

The Russian Government’s Coordination Council extensively uses DSS solutions based on mathematical modeling and machine learning methods, in particular, for prediction and regulation of consumer behavior. For example, prompt data analysis has allowed the federal government to prevent food shortages during the pandemic amidst a roaring demand for goods. Using reliable data obtained from various sources, the government provided assistance to private retail chains in coordinating prompt deliveries of goods to prevent deterioration in quality of life.

The glaring examples include the case of the Republic of Tatarstan, which implemented a pilot project to collect environmental data from several sources in 2020. The operations center obtained air quality indicators from the Ministry of Environment as well as from systems and devices installed at major strategic enterprises. Data updates allowed to promptly detect the sources of pollution and take respond measures to prevent major incidents and citizens’ complaints.

These examples show that operations center analysts have learned to detect data coherence as well as obvious dependencies. The next step is entrusting AI with handling unstructured data and implicit dependencies.

Step 4: Security and privacy

Many DSS developers claim that their products provide tools that offer unhindered and equal access to any data, enabling data democratization. Yet, this approach cannot be applied to public administration.

I believe that DSS should use advanced instruments for differentiating access rights to models and data through visibility scope created for different users; these tools should preferably be certified by FSTEC.

Conclusions

For over 15 years, work has been underway in Russia to build, develop and implement operations centers that comply with requirements, state standards and numerous regulatory documents. Yet, so far only certain constituent entities have advanced DSS able to predict an exact flood location, create the best transport route during roadwork, and say how fast certain socio-economic developments will occur. Naturally, regional governments that have successfully implemented the data-driven approach in public administration are deservedly at the top of rankings, showing a high gross regional product.

By Vasily Morozov, Head of Business Development, Innostage

Previous ArticleNext Article