Search results

    Search results

    Show all results for ""
    Can not find any results or suggestions for "."

    Search tips

    • Make sure there are no spelling errors
    • Try different search terms or synonyms
    • Narrow your search for more hits

    How can we help?

    Contact Us

    Find Employees

    University of Skövde, link to startpage

    Search results

      Search results

      Show all results for ""
      Can not find any results or suggestions for "."

      Search tips

      • Make sure there are no spelling errors
      • Try different search terms or synonyms
      • Narrow your search for more hits

      How can we help?

      Contact Us

      Find Employees

      University of Skövde, link to startpage

      Big data analytics

      Big data analytics

      Decisions require insights into the current and future business environment. Leveraging on the abundance of internal and external data sources, we can build analytics models to provide a rich data-driven objective view, supplementing domain expertise. Accounting for diverse streams of information results in a more holistic understanding of our business and market, leading to more sustainable operations and decisions, reducing waste, identifying new opportunities, and achieving informed, scalable, and automatable solutions.

       

      Key aspects of big data analytics are the predictive and prescriptive modeling. We use data to generate reliable predictions, understand the associated uncertainties, communicate these effectively to stakeholders, and translate them into actions. A critical aspect is the implementation of decision-relevant key performance indicators to monitor and improve processes, models, predictions, and insights.

      Predictive modeling is at the core of companies

      The different functions and planning efforts within a corporate environment require a multitude of different forecasts. Operations depend on a large number of, ideally automated, shorter-term forecasts, for instance, to support demand planning, inventory management, production planning, or predictive maintenance. Similarly, services depend on accurate forecasts, such as managing customer support, after-sales, and call center operations.

      Tactical decisions, such as budgeting, capacity planning, hiring and training staff, require mid to long-term forecasts, incorporating not only internal data sources but also external, such as macroeconomic and market variables. Likewise, managing product life cycles can incorporate the online activities and discourse of existing and potential customers to optimize product launches and their competitiveness.

      At a strategic level, the incorporation of scenarios and domain expertise are fundamental. Yet, experts can be biased and can struggle to incorporate multiple data sources in a balanced and effective way. Data-driven modeling can help to provide insights, inform on the likelihood of different scenarios, and identify leading indicators that shape the business environment.

      In all these, capturing and communicating the uncertainty contained in forecasts is a critical aspect to managing the inherent risk in decision making and associated costs.

      A predictive logic for understanding business reality

      Eventually, this brings a change in the way we perceive business reality. An appreciation of what is forecastable, actionable, and eventually under our control, can help in better utilization of resources, but also reaching to more data-driven and optimal decisions. An appreciation of the inherent uncertainties and their drivers can prepare an organization to deal with unprecedented disruptions and changes in the market. Ultimately, it provides a competitive advantage, leading to better, scalable, and trustworthy decisions, preparing us to meet future challenges.

      Responsible for the domain area

      Professor of Informatics

      Published: 3/14/2022
      Edited: 3/14/2022
      Responsible: webmaster@his.se