The use of machine learning to generate quick savings and gain in price consistency

The purpose of this article is to share our experience in the use of machine learning to generate quick saving opportunities and homogenize purchased pricing across business units / regions in large groups.

  Context and challenges within a multi-BU group

It could rapidly become cumbersome for a large group and its supply management team to ensure consistency of prices on a specific category.

  • check
    Why pay $5.6 dollars for an extruded tube of 10cm long and 2mm diameter and $20 for 20cm and 2.5mm diameter tube?
  • check
    Or a certain price for a red product and double for its blue version?

Those gaps can be normal but need to be controlled and understood so that suppliers or R&D can be challenged.

Even with a corporate organization and systems, Business Units are often independent or even worse, siloed. They could implicitly refuse to play it global because they need to go fast and stay independent. But, beyond organization & cultural behaviors, main barriers are technical & IT related:

  • check
    Products have different specifications & characteristics making the price comparison and alignment difficult
  • check
    There is no common repository combining technical & economical information
  • check
    History of pricing is rarely reviewed
  • check
    ERP system not alerting on pricing gaps
  • check
    Tail pricing is not properly controlled
  • check
    Pricing revision routines can be different across the group
  • check
    Multi contract with similar suppliers
  • check
    Suppliers apply a different cost of doing business per BU / Region

For sure, the traditional should cost approach and solutions can help solve these issues.

However, they are often complex to implement on a large scale, because they require a lot of manufacturing process related information, strong technical expertise and deep & time-consuming cooperation with suppliers (open book policy is mandatory).


 Refer to our article “Supply Chain optimization: a total cost approach

  Value proposition

Many firms have developed an innovative & effective approach combining advanced analytics and predictive algorithms (derived from Artificial Intelligence) to generate quick savings opportunities, only by processing your existing data.

The methodology is based on 5 pillars:

  • check
    Create a cost model based on shared cost drivers considering BU/Region specifics
  • check
    Utilize benchmarking to identify gaps across BU/Regions and optimized sourcing strategies
  • check
    Identify overpriced products and quantify savings opportunities
  • check
    Utilize data mining to determine optimization levers and prepare arguments for supplier negotiations
  • check
    Specify the management systems & required needs to sustain the process 

The following benefits have been observed:

  • check
    Identify savings opportunities (negotiation, VAVE, resourcing)
  • check
    Generate quick wins through “analytics based” negotiations with suppliers
  • check
    Control supplier quotes on new projects
  • check
    Improve Sourcing Strategies through optimized supplier allocation per cluster
  • check
    Drive continuous improvement through a robust cost model, increased skills, and improved cross-functional collaboration

  Machine learning based costing solutions

Machine learning based costing solution which estimates the price of a new product or service by processing current/historical data through a sophisticated algorithm, such as “Random Forrest”. Random Forrest is a nonparametric statistical method that performs learning on multiple decision trees driven on slightly different subsets of data generated by Bootstrap techniques (Ref. Breiman, L., Random Forests. Machine Learning. 45, 5-32 (2001)).

This type of methods allows to estimate the price of a product/service based on pre-identified parameters called “cost-drivers”. The estimation is very quick and accurate (30% of increased accuracy in comparison to traditional statistical methods).

The main advantages of machine learning based costing compared to traditional methods are:

  • check
    The user does not need to be an expert of the manufacturing process of the product. The estimate is based exclusively on the product characteristics (“cost drivers”) which are information you have access to internally (vs. asking to your supplier)
  • check
    It can mix an infinite number of cost drivers, those cost drivers can be continuous or discrete, technical (weight, function, color, raw material type, …) or commercial (country, volumes, supplier…)
  • check
    Ability to process databases for which the number of variables largely exceeds the number of observations
  • check
    Ability to identify and weight automatically the most important parameters, and therefore the cost drivers that impact the most
  • check
    Ability to interpret results
  • check
    Ability to manage missing values / incomplete database

For all these reasons machine learning based software helps create a very robust and ready-to-use costing model.

Beyond the above, machine learning based costing solution is processing all current purchasing prices and identifying inconsistencies / gaps versus estimates, which makes it easy to identify savings opportunities, including negotiations with suppliers.

Finally, some solutions have integrated benchmark functionalities which allow to compare each BU/Region for a specific category (even through products that have different design & characteristics).

They are increasingly utilized in industry and therefore is, every day, adding external benchmark knowledge for each commodity (IP and confidentialities being respected).

This allows to create benchmarking communities and share further (life sciences, automotive…)..

  Examples

Advantages

Limitations

Examples of software

Analytical Model

  • Explanatory and operations- centered model
  • “Best Landed Cost” Estimation and Target Price Definition
  • Allows to optimize the prices in production and to control the plans of progresses suppliers
  • Difficulty accessing process references and maintaining them over time
  • Intrusive approach towards suppliers
  • Expert model with little diffuse
  • Timeout for setting and performing encryption
  • Precision ?
  • Siemens PLM
  • A Priori
  • Facton
  • Statistical Parametric Model

    • Easy and quick to use
    • Estimated price coherence, and accuracy (conditional)
    • Non-intrusive approach to suppliers
    • Product and service applications
    • Very relevant in the upstream phases of the life cycle and for the analyses of coherence
    • Requires minimal data and quality history
    • Model that is not very "explanatory" to moderate supplier progress plans
    • Less relevant model for setting target prices and "Best Landed Cost"
    • Difficulty in modeling qualitative parameters
  • Seer
  • EstimFEC
  • Non-Parametric Statistical Model

    « Random Forests »

    • Easy and quick to use
    • Consistency of the estimated price, and precision increased by 30% compared to parametric models (conditional)
    • Non-intrusive approach to suppliers
    • Product and service applications
    • Very relevant in the upstream phases of the life cycle
    • Relevant also in the downstream phases for the analysis of price coherences and the identification of opportunities thanks to the explanatory properties of the forests
    • Integrates a lot of cost drivers, including qualitative ones
    • Detects technological breakthroughs
    • Prioritizes cost drivers
    • Manages missing values and can work with a limited sample
  • Model less relevant for setting target prices and "Best Landed Cost"
  • easyKost