Billions of dollars are tied up with maintaining stock norms in the Pharmaceutical industry. Nevertheless, Pharmaceutical companies are reluctant to cut inventory in contrast to FMCG companies, due to higher margins and the lifesaving aspects of certain drugs calling for higher levels of availability. However, there is a significant opportunity to reduce their inventory through optimized function in the supply chain which is supported by better planning

Artificial Intelligence is changing the landscape of decision-making in business. The rapid growth of technology that enhances processing speeds and storage architecture paves novel ways of autonomations, where we can train machines to assist in decision making.

Let’s take a typical ‘Order to Distribution’-process.

Fig 1: The Order to Distribution process

A typical company would want to maintain optimal level of inventory, i.e. No Out of stock or Over stock scenarios. The only touch point where the company has the liberty to influence that, is at the time of order placement.

When placing an order, the following factors need to be considered;

Fig 2: Order Placement Consideration

Demand forecast and the lead-time are the main uncertainties that needs to be captured at SKU level for the entire product portfolio.

To understand the complexity of this, lets consider the factors that determines the stock level of a basic flu drug to be maintained by the pharmaceutical company in your locality. Let’s call it brand X. The most likely factors that would influence the inventory level of a flu med includes a variety of variables; seasonality (seasonality of the drug sales in past years would give an indication of this), its-own and competitor’s promotions, economy related factors such as inflation and NCPI(people may opt for home remedies as opposed to western drugs during economically-challenged times), climate related factors such as rainfall, temperature and humidity, logistical factors such as lead time, custom clearance, and other inherent factors such as expiry dates of existing stocks are only a few out of the plethora of factors that would influence the inventory level of a flu drug.

Most companies lack the in-house capability and expertise to capture all the fluctuations and to quantify the factors of demand and lead time uncertainty. As a result, they end up maintaining excessive amounts of stock to avoid ‘stock outs’.

Fig 3: Common Pain-points and Repercussions

Maintaining an optimal level of inventory is important to a firm, both financially and operationally. From a financial viewpoint, efficient inventory management enhances profits by reducing the cost of procured pharmaceutical products and associated operational expenses. In addition, cash flow will improve with the savings on purchasing.

Building and refining an optimization model to capture all these factors is a highly complex task that requires a combination of domain and data science expertise and weeks/months of effort. This is what led to the development of Linear Squared’s Forecasting and Inventory Optimization Product. By leveraging the power of Artificial Intelligence and Machine Learning, this revolutionary product can provide the optimal inventory plan through a 4-step process

  1. Based on the forecast for a given SKU the algorithm will look at all possible demand variations and lead time variations for the given time period and evaluate the quantified effect of demand variation. (This is a key parameter for order plan generation.)

  2. Next, inventory simulation will be carried out based on the Demand, Current and Transit Inventory, Lead Time, Out of Stock risk % and other order specific details. Algorithm would take the variations in each of the attributes (taking all possible permutations) and generate order schedules for each permutation.

  3. Once the order schedules are generated, the schedule with the minimum inventory cost for the entire period in concern will be considered as the Optimal Order Schedule.

  4. In addition to generating the optimum order plan, the solution also allows stakeholders to perform scenario comparisons. For example, suppose the company plans to change their regular manufacturing plant to a different destination which may affect the lead time or suppose there’s a newly imposed drug regulation that may affect the import license registration.  

The product has the necessary provisions to accommodate such eventualities. It allows the stakeholders to observe the quantified effect the external factors have on demand projection and order schedule. For example, any effect on the order plan from exchange rate fluctuations would be easily calculated.

Fig 4: Key Outcomes Of Linear Squared’s Inventory Optimization Product

With an optimal inventory planned at each SKU level, wholesalers can forgo general company-wide policies on inventory management that typically treat all drugs to follow similar demand variations and delivery lead time variations. This granular focus at each SKU level would result in only holding inventory at required levels, avoiding unnecessarily buffers previously maintained due to generalized policies. What’s more, algorithms are self-calibrated with real time data to generate the most optimal and up-to-date order plan.

The result? Release of tied up working capital, lower risk of stock outs, lower holding costs, lower wastage due to expiry and overall efficiency through effective management of operations!

Sounds intriguing? Drop us an email today with your contact details and register for a FREE POC to demonstrate the impact of this remarkable product on your organization.

Feature Image Credits: https://www.contractpharma.com/