Nowadays, many companies face difficulties in decision-making processes, as, due to severe market competition and COVID-influenced recessed economies, the cost of mistakes becomes incredibly high. To offset the possible negative implications of wrong decisions on the company, more and more managers rely on statistical data to help them make the right decision. The analysis of statistical information often becomes a critical factor in decision-making processes. Before making decisions, many companies use the calculation of probabilities which often becomes an integral part of developing companies’ strategies. This paper discusses the use of probability in the decision-making processes of an ORGAMAN company as described in the article “Decision making in optimizing a product of a small-scale industry: a Bayesian analysis approach.” Moreover, it suggests some ways to improve the use of probability methods to tackle the problems faced by the company.
ORGAMAN company is a small-scale producer of organic fertilizers based in Jorhat District of Assam, India. The company is situated in a district where most of the population is involved in farming; the closeness to the sales market stipulates great demand for its products. The company has been operating on the local market for a number of years and is well-known by the customers (Biswas, 2018). However, the fertilizers produced by ORGAMAN are not unique; recently, the company has faced tough competition from other fertilizer-producing companies in the region. Trying to avoid the decline in sales, ORGAMAN looks into the sales of its well-known product “MUKTA,” which is organic manure (Bharali et al., 2018). The managers have to decide how much “MUKTA” they need to produce to get maximum profit, considering the expected demand. To do this, they use statistics from previous periods on the quantity of “MUKTA” production and the previously shown demand.
First, the table was drawn, which shows the pay-off for the “MUKTA” against the demand for this organic manure. In this prior analysis, two factors have been analyzed – Expected Monetary Value (EMV) and Expected Opportunity Loss (EOL) (Bharali et al., 2018). Using the joint probability approach, from this conditional pay-off table, for-profit managers have found that high production is preferable to lower output as it has greater EMV with prior probabilities (Bharali et al., 2018). Thus, the optimum quantity of product to be produced has been found to be 350 MT; the expected profit for this amount of product is Rs.675000 (Bharali et al., 2018). However, this figure does not consider possible further research in the field to get better results.
The calculations of joint probabilities of research value/additional payments for research show that, for ORGAMAN’s research to be beneficial, the company will have to pay Rs.50000 (Bharali et al., 2018). In the posterior analysis, calculations of the conditional probability of high demand/low production variant are done. These calculations are instrumental in finding join, marginal, and revised (posterior) probabilities. Then EMV with marginal and posterior probabilities has been calculated (Bharali et al., 2018). The results showed that if the company conducts research and gets new information, the expected profit will rise from Rs.675000 to 675364 (Bharali et al., 2018, p. 1867). The next step has been to determine the best variant considering prior and posterior probabilities.
To get the best option, the managers needed to see how much money they may spend on actual research. To do this, they calculated the Expected Value of Sampling Information (EVSI) (Bharali et al., 2018). The managers deduced EMV with prior probabilities from EMV with posterior and marginal probabilities; this figure has been found to be Rs. 364 (Bharali et al., 2018). This figure shows how much money the company may pay for conducting further research in the field. Then this figure has been compared to the amount of money needed to conduct research found in the prior analysis, which was Rs. 50000 (Bharali et al., 2018). The actual sum of money needed for the research is much higher than the one found optimal by the company’s managers. At the same time, the difference in profit in prior and posterior probability analysis is a meager one (Bharali et al., 2018). So, the managers decided to adopt a prior analysis strategy that featured high production without research.
The calculation of probabilities helped the managers adopt the right decision-making strategy regarding its “MUKTA” product. I recommend the company to calculate joint probabilities of research value against additional payments for research for its other products, especially those with innovative character. “MUKTA” product has been on the market for a long time, and this fact in itself shows that the possibilities of research here are limited and very costly. However, there are extensive possibilities for innovations and elaborating new products in the fertilizing industry that would be more effective and ecologically friendly. I believe that probability analysis would come up with different results if the products analyzed were more modern. As for the types of analysis used, I would recommend joint and marginal probability analyses, as these types show the correlation of properties against each other and allow managers to choose the best decision-making strategy.
References
Bharali, S., Patowary, A. N., & Hazarika, J. Decision making in optimizing a product of a small- scale industry: A Bayesian analysis approach. International Journal of Advanced Engineering, Management and Science, 2(11), 1865-1869.
Biswas, B. (2018). Impact of capital structure on profitability: A comparative study on some select public and private fertilizer companies in India. Sankalpa, 8(1), 40-47.