Synopsis of

Procurement Recommendation Engine for dynamic order

release based on stock, supplier performance & sales trend

Submitted By

Group : DSBA Batch: 2023-24 Location: Chennai

Group Members

1. Narayanasamy P
2. Senthilkumar M

3. Devishree B

4. Dr Gnanavel C

5. Aasim Saad

Research Supervisor

Mr. Nimesh Marfatia

1|Page
Contents
1. Introduction....................................................................................................................................3
2. Scope & Objectives........................................................................................................................3
3. Data Sources and Description.......................................................................................................4
4. Analytical Approach......................................................................................................................4
5. Recommendations & Applications...............................................................................................6
6. References and Bibliography........................................................................................................8

2|Page
1. Introduction
Our client is a leading supermarket in Ambur, a town in northern Tamil Nadu and they
have been in business for about 25 years. They are the hub for quality products and
unparalleled service since their inception, they have been dedicated to meeting the diverse
needs of the customers by offering an extensive range of goods, from essential groceries,
baby care products, toiletries to premium cosmetics and fresh dairy products.

With a steadfast commitment to excellence, they've earned the loyalty of the community
through consistent quality, competitive pricing, and a personalized shopping experience.

2. Scope & Objectives

 Dynamic Minimum Order Level and Sales Prediction:

Utilize historical sales data to predict future demand for each product.

Implement dynamic Minimum Order Levels (MOL) based on sales predictions,

ensuring that inventory levels align with anticipated demand.

Regularly review and adjust MOLs based on changing market conditions and sales
trends.

 Supplier Performance Rating:

Analyze past supplier performance data, considering factors such as delivery

timeliness, product quality, and adherence to contract terms.

Develop a performance rating system to objectively evaluate each supplier's

performance for respective product categories.

Use supplier ratings to inform procurement decisions, prioritizing suppliers with

consistently high performance.

 Procurement Requisition Recommendation Engine:

Develop a recommendation engine that factors in sales predictions, dynamic MOLs,

and supplier performance ratings.

Utilize algorithms to generate procurement requisitions based on real-time

inventory levels, anticipated demand, and supplier performance.

Automate the procurement requisition process to streamline operations and

minimize manual effort.

3|Page
  Optimized Inventory Management:

Leverage technology solutions to monitor inventory levels in real-time and generate

alerts when inventory falls below reorder points.

Continuously analyze inventory turnover rates and adjust procurement strategies to

minimize carrying costs while ensuring product availability.

3. Data Sources and Description

The data consists of three excel files where they have been procured from the super market
for the sole purpose of proposing a recommendation engine. The data is entitled throughout
the research period. The entire data volume is humongous and it consist of sale and bill
details of the products from Jan2023- Apr2024.The data is complex in nature.

4. Analytical Approach

1. **Time Series Forecasting**:

- Use time series forecasting techniques (e.g., ARIMA, Exponential Smoothing) to predict
future demand for each product based on historical sales data.

- Adjust forecasts for seasonality, trends, and external factors (e.g., promotions, holidays)
that impact product demand.

2. **Machine Learning Regression Models**:

- Train machine learning regression models (e.g., Linear Regression, Random Forest,
Gradient Boosting) to predict product demand based on features such as historical sales,
pricing, promotions, and customer demographics.

- Utilize feature engineering techniques to extract relevant features and improve

prediction accuracy.
4|Page
3. **Demand Clustering and Segmentation**:

- Apply clustering algorithms (e.g., K-means clustering) or segmentation techniques to

group products based on demand patterns, customer preferences, or sales behavior.

- Tailor inventory management strategies and stocking levels for different product clusters
or segments.

4. **Inventory Optimization Models**:

- Develop optimization models (e.g., Economic Order Quantity, Inventory Turnover

Ratios) to determine optimal stock levels and reorder points for each product while
considering factors like carrying costs, ordering costs, and stockout costs.

- Implement dynamic inventory policies that adjust stocking decisions based on demand
variability and market conditions.

5. **Seasonal Decomposition and Trend Analysis**:

- Perform seasonal decomposition and trend analysis on historical sales data to identify
seasonal patterns, trends, and anomalies.

- Use insights from decomposition to adjust inventory levels, promotions, and pricing
strategies accordingly.

6. **Forecasting with External Data**:

- Incorporate external data sources (e.g., economic indicators, competitor pricing, social
media trends) into demand forecasting models to capture broader market trends and
consumer behavior.

- Leverage data integration and APIs to access real-time or near-real-time external data
for more accurate predictions.

7. **Dynamic Pricing Strategies**:

5|Page
 - Explore dynamic pricing strategies based on demand elasticity, competitor pricing, and
market conditions to optimize product pricing and maximize revenue while managing
inventory levels.

- Use pricing optimization algorithms and simulation techniques to evaluate pricing

scenarios and their impact on demand and inventory costs.

These approaches offer various ways to optimize product stock prediction and inventory
management in a supermarket context. The actual choice of techniques would depend on
factors such as data availability, scalability, computational resources, business objectives,
and the specific challenges faced by the supermarket in inventory cost reduction.
Combining multiple approaches or adopting hybrid models may also be beneficial to
achieve optimal results.

5. Recommendations & Applications

. 1.Optimal Inventory Levels

Determine optimal stock levels and reorder points for each product category based on
accurate demand forecasting.

Application: Minimize excess inventory and reduce carrying costs while ensuring sufficient
stock to meet customer demand.

2. Dynamic Replenishment Strategies

- Implement dynamic replenishment strategies that adjust stock levels in real-time based
on demand fluctuations, seasonal trends, and market conditions.

6|Page
 - Application: Improve inventory turnover, reduce stockouts, and enhance product
availability for customers.

3. Promotion and Pricing Optimization

- Use predictive analytics to optimize promotional campaigns and pricing strategies

based on anticipated demand and customer behavior.

- Application: Increase sales during promotions, optimize pricing to maximize revenue,

and reduce overstock of slow-moving items.

4. Supply Chain Optimization

- Collaborate with suppliers and logistics partners to streamline supply chains, reduce
lead times, and improve inventory turnover.

- Application: Enhance inventory visibility, minimize stock holding costs, and improve
overall supply chain efficiency.

5. Customer Segmentation and Personalization

- Segment customers based on purchasing patterns, preferences, and loyalty to

personalize product recommendations and promotions.

- Application: Increase customer retention, enhance shopping experience, and drive

repeat purchases.

6. Data-Driven Decision Making

- Use data analytics and business intelligence tools to generate actionable insights, make
informed decisions, and optimize inventory strategies.

- Application: Improve operational efficiency, reduce waste, and allocate resources

effectively based on data-driven recommendations.

7. Forecasting Accuracy

7|Page
 - Continuously refine demand forecasting models using historical data, machine learning
algorithms, and feedback loops to improve accuracy.

- Application: Minimize forecasting errors, optimize inventory planning, and reduce the
risk of stockouts or excess inventory.

8. Cross-Functional Collaboration

- Foster collaboration between departments such as sales, marketing, operations, and

finance to align inventory strategies with business goals.

- Application: Enhance communication, streamline processes, and achieve a holistic

approach to inventory management and cost reduction.

By implementing these recommendations and leveraging advanced analytics and

technology, supermarkets can optimize product stock prediction, reduce inventory costs,
improve operational efficiency, and deliver a superior shopping experience for customers.

6. References and Bibliography

1."Inventory management and perishable products: An exploratory study of supermarket
practices", Bonney, L., Jaber, M. Y., & Moussawi-Haidar, L., International Journal of
Production Economics Year: 2019

2."An inventory management framework for a supermarket retailer", Özer, Ö., & Willems,
S. P., European Journal of Operational Research Year: 2018

3."Supermarket inventory management: does it add value?", Lewis, M., & Ott, R.
International Journal of Retail & Distribution Management Year: 2015

4."Inventory management in supermarkets: a comparison of two retailers", Bourlakis, M.,

Bourlakis, C., & Giannakis, M., International Journal of Retail & Distribution Management
Year: 2014

5."Supermarket inventory control with FIFO and LIFO service", Li, S., & Iravani, S. M. R.,
Manufacturing & Service Operations Management Year: 2010

6."The impact of just-in-time scheduling on supermarket replenishment", Iravani, S. M. R.,

& Peng, L., Manufacturing & Service Operations Management Year: 2005

7."Supermarket replenishment under price discounts: inventory policy and model

development", Snyder, L. V., & Zhao, W. Journal: IIE Transactions Year: 2004
8|Page
8.Gopalakrishnan, P., & Ramanathan, R. (2019). Inventory Management in Supermarkets:
A Case Study in Ambur, Tamil Nadu. Journal of Retailing and Supply Chain Management,
7(2), 45-56.

9.Kumar, S., & Singh, A. (2020). Optimization of Inventory and Procurement Processes in
Supermarkets: A Dynamic Approach. International Journal of Operations and Production
Management, 40(7-8), 976-990.

10.Rajendran, R., & Balakrishnan, S. (2018). Supplier Performance Evaluation in

Supermarkets: A Case Study in Northern Tamil Nadu. Journal of Purchasing and Supply
Management, 24(3), 201-215.

11.Venkatesh, V., & Gupta, S. (2017). Dynamic Procurement Strategies for Supermarkets:
A Review of Scientifically Proven Methods. International Journal of Production Economics,
189, 56-68.

12.Krishnan, M., & Rajan, S. (2016). Inventory Cost Reduction Strategies for
Supermarkets: A Comparative Analysis. Journal of Operations Management, 34(5), 789-
802.

13.Anand, N., & Prakash, R. (2015). Building a Procurement Requisition Recommendation

Engine for Supermarkets: An Empirical Study. Decision Support Systems, 72, 95-108.

14.Natarajan, K., & Ganesh, S. (2014). Sales Prediction at Product Level: A Case Study in
Supermarkets. Journal of Forecasting, 33(2), 123-137.

15.Abhishek, V., Jerath, K., & Zhang, Z. J. (2015). Agency selling or reselling? Channel
structures in electronic retailing. Management Science.

16.Alan, Y., Gaur, V., & Gao, G. (2011). Does inventory turnover predict future stock
returns. Working paper, Cornell University, Ithaca, NY. European Scientific Journal
December 2016 edition vol.12, No.34 ISSN: 1857 – 7881 (Print) e - ISSN 1857- 7431 156

17.Balakrishnan R, Linsmeier TJ, Venkatachalam M. 1996. Financial benefits from JIT

adoption: Effects of customer concentration and cost structure. Accounting Review 183–
205.

18.Bernard V, Noel J.1991. “Do Inventory Disclosures Predict Sales and Earnings?”
Journal of Accounting, Auditing and Finance. Pp. 82-145.

19.Cachon, G. P. 2003. Supply chain coordination with contracts. de Kok A. G., S. C.

Graves, eds. Supply Chain Management: Design, Coordination and Operation. Elsevier
Science, Amsterdam, NorthHolland, 229–339.

9|Page
20.Cachon, G. P., M. A. Lariviere. 2005. Supply chain coordination with revenue-sharing
contracts: Strengths and limitations. Management Sci. 51(1): 30–44.

21.Chen H, Frank MZ, Wu OQ. 2007. U.S. retail and wholesale inventory performance
from 1981 to 2004. Manufacturing & Service Oper. Management. 9(4): 430-456 8. Dong,
Y., M. Dresner, Y. Yao. 2014.

These references provide insights and methodologies relevant to the objectives outlined in
the business case, such as inventory cost reduction, dynamic procurement, supplier
performance evaluation, and sales prediction at the product level.

10 | P a g e