METHOD WE USE

The Decision scientists work closely with the clients and create analytical roadmap by understanding the list of the problems and prioritising them. Once the problems are prioritised, our world class team look at each and every decision making situation from five different perspectives; Behavioural science, Business consultancy, Applied Mathematics, Technology & Program Management and follow a scientific and holistic approach to solve them.

BUSINESS PROBLEM (QUESTION) FRAMING

  • Obtain or receive problem statement and usability requirements
  • Identify stakeholders
  • Determine if the problem is amenable to an analytics solution
  • Refine the problem statement and delineate
  • Define an initial set of business benefits
  • Obtain stakeholder agreement on the problem

ANALYTICS PROBLEM FRAMING

  • Reformulate the problem statement as an analytics problem
  • Develop a proposed set of drivers and relationships to outputs
  • State the set of assumptions related to the problem
  • Define key metrics of success
  • Obtain stakeholder agreement

HYPOTHESIS GENERATION

  • Identify the factors
  • Hypothesis framing process
  • Hypothesis testing process
  • Insights generation process

EXPLORATORY DATA ANALYSIS

  • Acquire data
  • Harmonize, rescale, clean and share data
  • Identify relationships in the data
  • Document and report findings (e.g., insights, results, business performance)
  • Refine the business and analytics problem statements

METHODOLOGY (APPROACH) SELECTION

  • Identify available problem solving approaches (methods)
  • Select software tools
  • Test approaches (methods)
  • Select approaches (methods)

MODEL BUILDING

  • Identify model structures
  • Run and evaluate the models
  • Calibrate models and data?
  • Integrate the models
  • Document and communicate findings (including assumptions, limitations and constraints)

DEPLOYMENT

  • Perform business validation of the model
  • Deliver report with findings;
  • Create model, usability and system requirements for production
  • Deliver production model/system
  • Support deployment

MODEL LIFECYCLE MANAGEMENT

  • Document initial structure
  • Track model quality
  • Recalibrate and maintain the model
  • Support training activities
  • Evaluate the business benefit of the model over time