What are data science business cases

TCW consulting service data science

Application areas and use cases

Data science and AI for early warning systems in procurement

Better prepared for component shortages and supply chain risks through data science and artificial intelligence. This is how the TCW approach works ... [more]

Successful implementation of data science in purchasing

Increased efficiency and reduced costs through the use of data science in operational purchasing at a plant manufacturer. ... [more]

Increase in sales through the optimized use of sales and platform data

The group-wide integration, evaluation and use of sales data from different business units offers companies ... [more]

System monitoring and predictive maintenance through machine learning

Historical operating, failure and sensor data of machines and large systems can be analyzed in real time with machine learning ... [more]

Quality optimization in the shop floor through the use of big data analytics

In many companies, production areas have different quality levels. Using algorithm-based big data analyzes ... [more]

Smart inventory management with big data and artificial intelligence

By optimizing the inventory management, potential can be realized in numerous company areas ... [more]

Remote Business Intelligence as a reaction to the Corona crisis

Remote Business Intelligence means assistance with strategic decisions through tool-based data analysis ... [more]


Data science means systematizing learning from data

The vast amount and variety of existing and new data generated in the world today is unparalleled. With the right tools, patterns can be recognized which can be used to optimize processes and resource allocation.

How does TCW support the introduction of data science?

We help our customers to identify valuable and meaningful insights from their data and to convert them into competitive advantages. TCW supports you in the identification of meaningful fields of application, in the preparation and consolidation of data up to the introduction and organizational anchoring of data science in the company. We help you to make fact-based decisions with the help of data and thus to become an analytically oriented company.

We support customers from the development of use cases through the development of prototypes to the company-wide rollout.

We link

  • Empirical knowledge from different industries,
  • a benchmark database with reference case studies of successfully implemented applications,
  • pragmatic analytics tools, which also make distributed and incomplete data stocks usable and
  • State-of-the-art analysis software and proven algorithms in a holistic concept.

Here we act as technology-independent consultant and Implementation supporter. We are convinced that a valuable consulting approach only arises when the data analysis is combined with human experience, because then the right conclusions can be drawn from it.

What does the TCW process model look like?

The TCW uses a four-step process that ensures measurable results:

1. Use case hypothesis
  • Analysis and benchmarking of the corporate function to be considered,
  • Identification of pain points and value levers in the value chain,
  • Identification of possible use cases,
  • Deriving cause / effect analysis to derive data science hypotheses
2. Data exploration
  • Analysis of data quality and data availability
  • Consolidation of data from various data sources (no perfect data quality necessary) and
  • Data mapping and data cleansing
3. Business case
  • Evaluation of the added value of various data science models for the company
  • Estimation of the investment costs and the implementation time
  • Cost / benefit analysis to evaluate various fields of application
4. Prototype
  • Pattern recognition in the data lake to identify patterns and abnormalities
  • Testing of different algorithms and algorithm selection
  • Design of a functional data science model and application tools
5. Piloting
  • Introduction of the prototype in the corresponding company area and application under real conditions
  • Refinement of the tool and further development of the model based on the lessons learned
6. Rollout
  • Introduction and integration of the model into company processes
  • Implementation support and training of employees
  • Transfer of know-how to other fields of application

Which competence is the unique selling point of the TCW?

Our teams are not just made up of mathematicians or IT experts. Our data scientists work on the Interface between data analytics and business administration, that means we act as intelligent translators and support the companies in the development of concepts that stand out realizable potential distinguish.

“The data scientists at TCW have the competence to express operational or strategic management decisions through analytical problems. This is made possible by a double qualification in technology and business administration. "

The data scientists bring experience and analytical know-how with them in order to quickly identify which particularities in the company area have to be taken into account and, above all, where the value levers are! That means, for example, knowing how the sales process works. What are the success factors in logistics? Which parameters come into question if I want to carry out a statistical analysis of the quality in the production network? The pairing of human experience and algorithmic analysis power is the basic requirement for generating new knowledge.

Our interdisciplinary teams

  • consisting of economists,
  • Data Scientists and
  • Engineers

link technical understanding, industry and process experience and analytical skills in the area of ​​big data analytics.

We use a wide range of tried and tested algorithms and software tools from the areas of mathematical optimization, statistical processes, non-linear models such as decision trees, and even our deep learning (deep neural networks). The TCW concept enables companies to generate real added value from data and to identify opportunities to increase sales and reduce costs. We build on experience from successful reference projects.

Further use cases in the area of ​​data science

With advanced analytics, business problems can be uncovered that are hidden in the jungle of mass data, but can be brought to light with the tools of statistical data analysis. This approach can be applied in all areas of the value chain. We have presented some very prominent applications at this point:

  • Increased efficiency in operational purchasing
    Big data analyzes make it possible to establish connections between purchasing decisions (e.g. BANF releases) and the combinations of characteristics of the BANF. By clustering BANFs according to the criteria that led to a release in the past, the BANF can be released (or triggered another action) without manual checking if it falls into the corresponding category.
  • Pricing models for hybrid service packages
    The aim is to develop market and region-specific pricing models that enable maximum market absorption. By clustering regional customer requirements and recognizing patterns, hybrid service bundles can be offered and an optimal price strategy can be developed. The aim is to optimally take into account characteristics that are decisive for purchasing in the pricing strategy.
  • Cross-plant capacity simulation
    The cross-plant simulation of available capacities, taking into account the technological competencies, is often difficult in companies due to the complex interrelationships. Big data analyzes make it possible to standardize and compare the multitude of capacity-relevant data in the individual plants. This enables forecast-based, optimized resource planning and a gain in efficiency to be achieved.
  • Optimization of productivity in production through cause / effect analyzes
    Regression big data analyzes can be used to determine which causes favor high productivity (e.g. first pass yield, throughput times, etc.). The aim is to analyze successful and less successful locations in a production network and to identify patterns that can increase productivity.
  • Accelerated product development time in the chemical industry
    New products are developed in the chemical industry in laboratory tests. Many influencing factors such as recipes or process parameters have an influence on the product quality. Laboratory tests can be significantly more efficient if the patterns between influencing factors (recipes, processes, etc.) and process results can be created using big data. This means a significant acceleration compared to the trial-and-error procedure.
  • Pricing of complex product portfolios
    Automated pricing of the variants with more precise consideration according to the specifics and optimal absorption of the market.
  • Optimization of material placement
    Determination of the packaging sizes taking into account all influencing factors from the point of view of assembly and logistics. Utilization optimization in pre-assembly areas taking into account all influencing factors.