CASE STUDY

TEA BLENDING OPTIMIZATION
 
Industry
A mind-boggling 1 million cups of tea are drunk every minute of every day – making tea the world’s 2nd most consumed beverage in the world, after water. Tea is more than just a product – it touches so many people, it has a heritage, offers opportunities, and is, indeed, a way of life. While formulating new tea recipes, sustainable sourcing is as fundamental a consideration, in blend development as is raw material cost or its seasonality.
The Indian Tea Industry has an important and special place in the Indian economy. Tea is India's primary beverage, with almost 85% of total households in the country consuming tea. India is the world's largest producer and consumer of tea, with India accounting for 27% of the world tea production. Goldstein Research analyst forecast that the India tea industry outlook is set to reach USD 1 billion by 2025, growing at a CAGR of 5.8% over the forecast period (2017-2025).
About the Client
The client is a large MNC who is among the leaders in the FMCG Industry. The tea business is an important part of its FMCG business It has multiple tea brands and many varieties of tea. The demand for its tea blends runs into more than 50,000 tonnes in India (21.2 share in 2018) where the solution was meant to be implemented. Maintaining the standardization of various aspects of the tea blends and yet optimizing the cost is complex and challenging.
 
Business Challenge
How to create a tea blend from a wide range of raw material, in the inventory which is inherently multi-characteristic and create tea blends that adhere to a variety of defined quality score parameters, with zero violation of rules of formulation and achieve cost saving.
 
 
Objectives
  • The median score of each quality parameter of every lot of 'tea blend', must be within the prescribed range.
    Grading Parameter Minimum Maximum
    Brightness xx.xx* xx.xx
    Colour xx.xx xx.xx
    Softness xx.xx xx.xx
    Flavour xx.xx xx.xx
    *some number
  • Each characteristic of every blend must conform to all minimum and maximum percentage of blend quantity, as prescribed in the Blend Formulation Document of the R&D department.
  • Any characteristic that is not permitted for a particular blend must result in the batch of raw material being excluded.
Restrictions
  • Availability of stock and its nature at the time of blend creation.
  • Splitting of bags of the raw material are to be minimized to the extent possible.
Optimization Objectives
  • Create blends from the raw material in inventory to get the least cost per kg.
  • Spread the cost benefit so as to avoid overconcentration of lesser cost in one or more blends.
  • Fulfil maximum demand as per the demand plan.
Blend Simulator Solution

We firstly investigated the challenges in blending in the tea industry based on meeting with members of the R&D, Procurement team. Subsequently, the team visited the Kolkata tea centre of the company to meet the tea blenders and tasters and finally a visit to the factory.
After studying the challenges and brainstorming on various approaches, we proposed a multi-objective model that would empower the blenders to create blend lots that is an optimised outcome. The solution, has resulted in the company being able to save money of a huge magnitude on a weekly basis. In the next phase, Price and Usage Trend analytic tools were added to the solution.


Blend names, Total quantity, Lot size.

Get current stock of 'Available' raw materials.

Hide 'Excluded' line items from blend.