Flipkart Case Study | Restaurant optimization

Question

Punjabi By Nature is going through a rough phase and is wasting 30-40 Kgs of biryani on some days of the week.

Management is thinking to find a method to sell the left over biryani or reduce the amount of biryani prepared.

You are the manager and you are supposed to come up with a methodology for the same

solved 0
Xtramous 1 year 10 Answers 1279 views Contributor 1

Answers ( 10 )

  1. by when we can get an answer to these questions? Also, any particular approach followed for these types of questions.

  2. First, on seeing the trend of sales of biryani, we can make a decision for the apt/ estimate the amount of Briyani we can sell in a day, given the weekday, for instance on Tuesday the demand for briyani would be lesser than the other days. The location of the restaurant would also influence the sales of briyani, the demographics of the areas within let say 2-3 km of the radius would influence the sales of briyani. We can come up with special discounts on briyani at the locations where we are facing the competition from the other vendors selling briyani.

  3. As a manager, I would follow below steps:
    1. Look out for trends by consolidating visitors history data. Depending on how old the restaurant is.
    2. Looking at the trend I would be interested to know like
    Which day customer’s inflow is more or less depending on weekdays/weekends/holidays/festival
    How much biryani is getting cooked/served on daily basis.
    3. We can reduce the quantity of cooked biryani on certain days.
    4. We can plan for offers or discounts accordingly.
    5. We can reuse the food items some or other way and prepare some other dish.
    6. Also my personal advise – if we can parcel and sell those in lesser rate to needy and poor people rather than wasting or throwing it away.

    Best answer
    • That is quite a good approach 🙂

      • Assumptions:-
        – PBN is a restaurant and not a home kitchen.
        – They cook biryani only once a day.

        Here is kind of a mindmap I created for this problems with my idea’s and suggestions

        – Reduce the amount cooked/ Sell leftover

        – Selling Leftovers
        – Leveraging food delivery apps
        – Exclusive Partnership with food delivery apps with good offers during closing hours
        – Free testers with other orders of delivery app
        – Leveraging footfall in restaurant
        – x% off for customers on donating food for the needy
        – Exclusive buy 1 get 1 on specific days of weeks where wastage is more historically
        – Marketing
        – selling them nearby competitor stores with publicity pamphlets

        – Estimating the cooking amount
        – Based on data
        – Reduced quality/ quantity of food/ Prices
        – Quality
        – Recent change in the cook for biryani
        – Customer reviews on popular food critic website
        – Quantity
        – Change in portion size
        – Prices
        – Change in prices from history
        – Competitors nearby
        – New competitors nearby or on delivery apps
        – Prices and offers compared to competitors
        – Change in customers eating habits
        – Footfall and ordering trend for all the competitors

        Historic data for the following columns can be asked for further analysis – (Order history, Cook information,
        Customer feedback from various websites, portion sizes per plate, price per plate, Competitors order history)

        Leveraging above columns, a suitable model can be trained giving a probabilistic estimate for the amount of biryani
        to be cooked everyday.

        – Non data based
        – X% off on reserving table or an order a day before
        – Cooking biryani twice a day instead of once allows more control on quantity cooked

  4. As a manager i would first try to get the desired data to find insights related to our sell and make strategy based on it,
    (1) would take the Data of amount of Biryani cooked on each weekdays
    (2) would study the data of sale of biryani on each weekdays(keeping in mind of the weekends& holidays)
    solutions that i will suggest-
    (1) will definitely ask to reduce the quantity of biryani on weekdays where we don’t see much sell like on Tuesdays and Thursdays(Since in India it is prevalent to not eat non-veg on this days)
    (2) Will collaborate with local Tiffin/Mess people so that i can sell them cooked biryani twice a week at a lower cost to minimize my losses due to wastage.
    (3) will provide special discounts on takeaways and online ordering to increase my sell.
    (4) Will do hardcore marketing of my product and offers.

  5. Restaurants usually predict how much food will be needed by estimation. I would like to use a data-driven approach to refine the estimation.
    1) As a manager, I would like to take a look at historical data and see how the trend has been over the years. Has PBN grown? Is there an increase in the number of customers and/or number of branches?
    2) Are there two branches opened nearby resulting in incorrect estimation and hence, the wastage? Or has a new competitor restaurant opened up nearby?
    3) I would see if I can observe a seasonality. The sales might be going up on days like public holidays or Sundays. The food quantity should be greater on these occasions. However, on Tuesdays, the quantity of non-veg biryani needed would be less (demographic info)
    4) When we have left-over biryani of a particular kind and the day is coming to an end, do we offer more discounts compared to our competitors to increase the sales? We can offer discounts on food delivery apps at the end of the day and this could influence people to place orders online
    5) On certain days when we have too much leftover food, we can keep our restaurant open for deliveries at night as there are limited good options available for employees working at night
    6) Can we use the biryani as a substitute to plain rice in some kind of thalis?
    7) We can put up a stall at snacks or dinner time in mess of some companies if permission is granted to finish the extra food
    8) If I do have some time and resources, I would like to plot a time-series graph of quantity required and forecast values for better results!
    9) If there is still extra biryani left I would like to offer it to my employees, cleaners, cooks and some poor people rather than wasting it

    Other than this, I would also take into account customer feedbacks and if any particular type of biryani is left over, I would like to know if it is because of the decrease in quality or a poor service from the restaurant.

  6. Sorry it a private answer.

  7. Data analysis can be done on the amount of biryani prepared each day and the amount of biryani sold.
    This can be done differently for special occasions such as when there is huge sale of biryani and some days for which there is no sale of biryani.
    This would help us to-
    – reduce the production of biryani on some days.
    – offer discount and special offers to increase the sale.

    Another approach is also to reuse the leftover items to prepare a new dish or if the life of the food products is more, then it can be used for the next day.

  8. Let me split the answer into Two:
    Understanding the problem:
    1) Check the historical data and segregate that into days in a week and time slots (Lunch 12-3, Dinner (7-11). This will give us an estimate of how much biriyani is sold per time slots in a week. If more business for lunch, maybe we can reduce the quantity cooked for dinner.
    2)Also, check online orders and discounts provided. Coz sometimes if there is an issue with payment or order confirmation in food aggregator like Swiggy, it will directly impact PBN’s business.
    3) We can also consolidate our regular customers (dine in and online) based on their order pattern. If there is a decline in that, we can plan a loyalty plan or spl discounts.
    Action plan:
    1) Customer feedback (dine in and online) is very critical. If the feedback is bad, we need to rectify that asap.
    2)Also, we need to check what are our most sold items and check if we can couple biriyani with that.
    3) If we can estimate and pre pack biryanis, we can reduce the TAT in a swiggy delivery.
    4)Corporate discounts or biryani mela (say during IPL) can help in attracting new customers and selling the biryani
    5) Late night delivery
    6) If we have data available,we should check our competitors.
    This way we can first estimate how much to cook and then promote our biryani

  9. Assumptions:-
    – PBN is a restaurant and not a home kitchen.
    – They cook biryani only once a day.

    Here is kind of a mindmap I created for this problems with my idea’s and suggestions

    – Reduce the amount cooked/ Sell leftover

    – Selling Leftovers
    – Leveraging food delivery apps
    – Exclusive Partnership with food delivery apps with good offers during closing hours
    – Free testers with other orders of delivery app
    – Leveraging footfall in restaurant
    – x% off for customers on donating food for the needy
    – Exclusive buy 1 get 1 on specific days of weeks where wastage is more historically
    – Marketing
    – selling them nearby competitor stores with publicity pamphlets

    – Estimating the cooking amount
    – Based on data
    – Reduced quality/ quantity of food/ Prices
    – Quality
    – Recent change in the cook for biryani
    – Customer reviews on popular food critic website
    – Quantity
    – Change in portion size
    – Prices
    – Change in prices from history
    – Competitors nearby
    – New competitors nearby or on delivery apps
    – Prices and offers compared to competitors
    – Change in customers eating habits
    – Footfall and ordering trend for all the competitors

    Historic data for the following columns can be asked for further analysis – (Order history, Cook information,
    Customer feedback from various websites, portion sizes per plate, price per plate, Competitors order history)

    Leveraging above columns, a suitable model can be trained giving a probabilistic estimate for the amount of biryani
    to be cooked everyday.

    – Non data based
    – X% off on reserving table or an order a day before
    – Cooking biryani twice a day instead of once allows more control on quantity cooked

  10. Assumptions:-

    – PBN is a restaurant and not a home kitchen.
    – They cook biryani only once a day.

    Here is kind of a mindmap I created for this problems with my idea’s and suggestions

    – Reduce the amount cooked/ Sell leftover
    – Selling Leftovers
    – Leveraging food delivery apps
    – Exclusive Partnership with food delivery apps with good offers during closing hours
    – Free testers with other orders of delivery app
    – Leveraging footfall in restaurant
    – x% off for customers on donating food for the needy
    – Exclusive buy 1 get 1 on specific days of weeks where wastage is more historically
    – Marketing
    – selling them nearby competitor stores with publicity pamphlets
    – Estimating the cooking amount
    – Based on data
    – Reduced quality/ quantity of food/ Prices
    – Quality
    – Recent change in the cook for biryani
    – Customer reviews on popular food critic website
    – Quantity
    – Change in portion size
    – Prices
    – Change in prices from history
    – Competitors nearby
    – New competitors nearby or on delivery apps
    – Prices and offers compared to competitors
    – Change in customers eating habits
    – Footfall and ordering trend for all the competitors

    Historic data for the following columns can be asked for further analysis – (Order history, Cook information,
    Customer feedback from various websites, portion sizes per plate, price per plate, Competitors order history)

    Leveraging above columns, a suitable model can be trained giving a probabilistic estimate for the amount of biryani
    to be cooked everyday.

    – Non data based
    – X% off on reserving table or an order a day before
    – Cooking biryani twice a day instead of once allows more control on quantity cooked

Leave an answer

Browse
Browse