*2.6. Data Analysis*

#### 2.6.1. Loss Rate

Several approaches for determining quantitative losses have been developed. The counting and weighing methods were used to assess the rates of maize loss during storage. This method is the most used in studies and produces better results.

The formula used by Pantenius [17] is as follows:

$$\%Losses = \frac{(E\*B) - (C\*D)}{(E\*A)} \* 100\tag{1}$$

where *A* is the total number of grains, *B* is the number of damaged grains, *C* is the number of healthy grains, *D* is the weight of damaged grains and E is the weight of healthy grains.

#### 2.6.2. Storage Costs

The cost of the maize storage encompassed both fixed and variable costs. Thus, the total cost of maize storage was obtained by summing up the fixed and variable costs. The fixed costs included the costs of the storage structures and small storage equipment, such as basins and baskets. Following Arouna et al. [20] the average monthly cost was calculated as:

$$E(j) = \frac{\mathbb{C} - R}{n} + [(\mathbb{C} - R) \* f + R] \* (q - 1) + \mathbb{C} \* r \tag{2}$$

where *j* is the type of storage structure or storage small equipment, *E*(*j*) is the monthly cost of the storage structure or small storage equipment *j*, *C* is the storage structure construction cost or purchase price, *R* is the residual value of the storage structure or small storage equipment, *n* is its useful lifespan, (*q* − 1) is the interest rate, *f* is the capital asset factor, and *r* is the repair or maintenance cost factor (coefficient). The capital asset factor is estimated using the formula as follows:

$$Capital asset factor = (Interestrate/100) \* maizeselling price \tag{3}$$

In the study area, the small equipment was mostly not repaired and used until it was thrown away. Thus *R* = 0 and *r* = 0 and Equation (2) is written as follows:

$$E(j) = \frac{\mathbb{C}}{n} + \mathbb{C} \* f \* (q - 1) \tag{4}$$

The variable costs of maize storage encompassed the costs associated with storage losses and other variable costs. The storage loss in monetary value, also called financial loss, represented the quantified sum of the quantitative losses occurring during storage with a given treatment. Thus, to calculate the quantitative loss, the quantity stored was multiplied by the loss rate. The financial loss was obtained by multiplying the quantitative loss by the average monthly maize selling price. The other variable costs comprised the labor costs associated with maize storage operations—including the dispatching and shelling costs of maize before storage in storage structures, the cost of loading and unloading the maize from the storage structure, the cost of the application of the conservation measure Actellic® Super and interest on the capital asset of the maize stock.

#### 2.6.3. Benefit–Cost Ratio (BCR)

The benefit–cost ratio (BCR) was estimated for each treatment.

It was obtained by dividing the flow of the present value of the benefits (benefits) by that of the present value of the cost following the formula of Gittinger [18].

$$\frac{B}{C} = \sum\_{t=1}^{n} \frac{B\_t}{(1+i)^{t-1}} \Big/ \sum\_{t=1}^{n} \frac{C\_t}{(1+i)^{t-1}} \tag{5}$$

where *Bt* is the monthly benefits, *Ct* is the monthly costs, *n* is the storage duration (in months), *i* is the interest rate (expected) and *t* is a given month.

#### 2.6.4. Break-Even Quantity

The break-even quantity was estimated for each treatment, and thus, the break-even quantity of the turnover and the break-even ratio were determined.

The break-even quantity in turnover (BREQ) was calculated using the following formula:

$$BREQ = \left( Fixedcosts\* \frac{Revenue}{Grossmargin} \right) \* 100 \tag{6}$$

The percentage of capacity used, also called the break-even ratio, indicates the percentage of production for which the gross margin covers the fixed costs. The risk increases as the capacity percentage increases; a low percentage (maximum = 1) gives a level of security

against unpredictable operating difficulties. Therefore, when the value of this capacity tends towards 1, there is a risk in which the producer may no longer be able to pay for the equipment used. This percentage is computed by the following formula:

$$\%capparity used = \left(\frac{BREQ}{Revenue}\right) \ast 100\tag{7}$$

The break-even quantity of maize to be stored in each storage and preservation technology to make the investment profitable is given by the following formula:

$$\text{Break} - \text{evenquality} = \% \text{capacity} \,\text{used} \,\ast \,\text{chemical capacity} \tag{8}$$

#### 2.6.5. Statistical Analysis

The Student–Newman–Keuls (SNK) ANOVA statistical test was used to test the difference between the paired means between the different storage and conservation technologies using SPSS Statistics software. These paired multiple comparison tests captured the difference between paired means and generated a matrix that informed the means of groups of significantly different storage and conservation technologies [21].

#### 2.6.6. Sensitivity Analysis

The sensitivity analysis consisted of varying the fixed costs and the selling prices of maize to assess the effect of the change in certain parameters (fixed costs and selling prices of maize) on the benefit–cost ratio and the threshold quantity for each of the technologies. This assessment allowed the impact of the different treatments implemented on the economic profitability of maize storage to be observed when these parameters changed for any reason.

#### **3. Results**

#### *3.1. Average Loss Rate Recorded for the Different Treatments Implemented*

The PICS bag with grain treatment and the PICS bag without grain treatment recorded fewer losses in the two communes during storage (Tables 1 and 2).

**Table 1.** Evolution of the average loss rates of the different treatments during storage in Savalou.


*p* > F probabilities are indicated by symbols: ns = no significant differences; \*\* significant differences at *p* < 0.05; \*\*\* significant differences at *p* < 0.01. For each column, values with the same letter indicate no significant differences at 5%; Source: Experimentation data, 2015, 2016 and 2017 (− ctrol means: without chemical conservation measure; + ctrol means: with chemical conservation measure).



*p* > F probabilities are indicated by symbols: ns = no significant differences; \*\* significant differences at *p* < 0.05; \*\*\* significant differences at *p* < 0.01. For each column, values with the same letter indicate no significant differences at 5%; Source: Experimentation data, 2015, 2016 and 2017 (− ctrol means: without chemical conservation measure; + ctrol means: with chemical conservation measure).

The statistical differences observed showed that a significant increase in the loss of dry matter was observed at the level of each treatment throughout the six months of storage in Boukoumbé (*p* < 0.0001). In Savalou, significant variations were observed during the first two months. However, in the last month of storage, in the commune of Savalou, the PICS bag with a chemical conservation measure (9.42 ± 4.64%), the polypropylene bag with a chemical conservation measure (10.56 ± 2.80%) and the PICS bag without a chemical conservation measure (11.71 ± 2.78%) were the three treatments that recorded fewer losses, while the improved clay granary (22.64 ± 7.21%) recorded the highest loss rate (Table 2). In Boukoumbé, the PICS bag with a chemical conservation measure (2.69 ± 0.77%), the polypropylene bag with a chemical conservation measure (4.02 ± 1.23%) and the metal silo with a chemical conservation measure (4.92 ± 1.36%) were the treatments that recorded significantly fewer losses in order of priority, while the improved clay granary without a chemical conservation measure (12.57 ± 3.68%) was the treatment that recorded the most maize loss during storage (F = 3.093, *p* < 0.0001) (Table 2).
