Prof. Amar Sapra to lead IIMB Chair of Excellence seminar on optimal inventory management for online retailing
Research seminar to be held on 4 November from 4-5 pm
30 October, 2024, Bengaluru: Prof. Amar Sapra, Chairperson, MBA-Executive Post Graduate Programme for Experienced Professionals (EPGP), and faculty of the Production and Operations Management area at IIM Bangalore will lead the IIMB Chair of Excellence Seminar on 4 November, from 4 to 5 pm, in the Central Pergola at IIMB. The session will be themed on ‘Please Customers or Prevent Wastage? Replenishment and Issuing Policy for a Perishable Product with age-sensitive demand’. Prof. Amar Sapra held the IIMB Chair of Excellence position from September 2021 to August 2024.
To register, click here: https://docs.google.com/forms/d/e/1FAIpQLSeZDtehO9bXfcFAO50MlBYVDJBmk7gVSfkRojQNx7IYrJdS0w/viewform
Abstract: An online retailer selling a perishable product can issue inventories of different lifetimes in her preferred sequence. This leads to a trade-off between customer satisfaction and outdating costs since customers want freshness whereas the retailer desires units be sold within their lifetime. We study this trade-off by considering joint replenishment and issuing policy for a perishable product with general lifetime. We develop a periodic review model over a finite horizon and model customers' sensitivity to the remaining lifetime of the unit received using a goodwill cost, which increases as the remaining lifetime decreases. Even though customer satisfaction is likely to be maximized by the last-in, first-out (LIFO) issuing policy and the outdating cost is likely to be minimized by the first-in, first-out (FIFO) issuing policy, neither policy may be optimal. In particular, when the product lifetime is equal to two periods, it is optimal to first issue new inventory till a threshold and then issue rest of the inventory in FIFO order. Although the value function is always concave, for longer product lifetimes, the optimal issuing policy can become complex. Consequently, we focus on bucket policies, in which inventory of each remaining is issued in a single installment. We show that for a subset of bucket policies, the value function is anti-multimodular. Furthermore, our experiments showcase that the profit of the best bucket policy is on average within 0.1% of the optimal profit. We also develop an easy-to-compute heuristic that determines a bucket policy as well as order quantity and show computationally that its performance is within 1.3% of the optimal profit on average. The heuristic also outperforms many replenishment and issuing policies that have been presented in the literature. Although optimal issuing policies that capture the trade-off between customer satisfaction and wastage due to expiry can be complex, simple, effective, and easy to implement issuing policies can be identified; the performance of these policies is close to optimal.
Prof. Amar Sapra to lead IIMB Chair of Excellence seminar on optimal inventory management for online retailing
Research seminar to be held on 4 November from 4-5 pm
30 October, 2024, Bengaluru: Prof. Amar Sapra, Chairperson, MBA-Executive Post Graduate Programme for Experienced Professionals (EPGP), and faculty of the Production and Operations Management area at IIM Bangalore will lead the IIMB Chair of Excellence Seminar on 4 November, from 4 to 5 pm, in the Central Pergola at IIMB. The session will be themed on ‘Please Customers or Prevent Wastage? Replenishment and Issuing Policy for a Perishable Product with age-sensitive demand’. Prof. Amar Sapra held the IIMB Chair of Excellence position from September 2021 to August 2024.
To register, click here: https://docs.google.com/forms/d/e/1FAIpQLSeZDtehO9bXfcFAO50MlBYVDJBmk7gVSfkRojQNx7IYrJdS0w/viewform
Abstract: An online retailer selling a perishable product can issue inventories of different lifetimes in her preferred sequence. This leads to a trade-off between customer satisfaction and outdating costs since customers want freshness whereas the retailer desires units be sold within their lifetime. We study this trade-off by considering joint replenishment and issuing policy for a perishable product with general lifetime. We develop a periodic review model over a finite horizon and model customers' sensitivity to the remaining lifetime of the unit received using a goodwill cost, which increases as the remaining lifetime decreases. Even though customer satisfaction is likely to be maximized by the last-in, first-out (LIFO) issuing policy and the outdating cost is likely to be minimized by the first-in, first-out (FIFO) issuing policy, neither policy may be optimal. In particular, when the product lifetime is equal to two periods, it is optimal to first issue new inventory till a threshold and then issue rest of the inventory in FIFO order. Although the value function is always concave, for longer product lifetimes, the optimal issuing policy can become complex. Consequently, we focus on bucket policies, in which inventory of each remaining is issued in a single installment. We show that for a subset of bucket policies, the value function is anti-multimodular. Furthermore, our experiments showcase that the profit of the best bucket policy is on average within 0.1% of the optimal profit. We also develop an easy-to-compute heuristic that determines a bucket policy as well as order quantity and show computationally that its performance is within 1.3% of the optimal profit on average. The heuristic also outperforms many replenishment and issuing policies that have been presented in the literature. Although optimal issuing policies that capture the trade-off between customer satisfaction and wastage due to expiry can be complex, simple, effective, and easy to implement issuing policies can be identified; the performance of these policies is close to optimal.