An incorrect assessment of the effort, dedication, and professionalism of soldiers and officers may have a direct impact on declining morale.
Imagine 120 Border Guard posts and military units guarding the borders. Some are located in the mountains, others very close to cities, and still others in remote areas, in swamps, or in the middle of a forest. Some struggle with frequent smuggling and illegal border crossings, while others are monotonous places where there is little activity. Some posts are new, while others are old. Some have modern vehicles, while others face significant equipment shortages. All these facilities must somehow be assessed, even though each is different. Each struggles with a different kind of problem, and all are staffed by people who are sensitive to how we assess them.
Assessing facilities that perform similar functions and operate under the same rules is crucial because it makes it possible to reward the best or influence improvement in the work of others. By comparing different facilities, one can identify factors that determine an improvement in quality or cause it to decline. Assessment is a double-edged sword. It can be a powerful stimulus affecting soldiers’ morale, but an unfair assessment can also reduce it drastically and adversely affect their motivation to act.
Let us therefore return to our example. We have different Border Guard observation posts that differ significantly in their level of hardship, location, and equipment. How, then, can their work be assessed fairly?
DEA-CCR Comparative Analysis
Data Envelopment Analysis—Charnes, Cooper, and Rhodes (DEA-CCR)—is a method used to assess the efficiency of Decision-Making Units (DMUs), such as enterprises, institutions, hospitals, or military units, that transform inputs (for example, resources and expenditures) into outputs (for example, products and services).
The term Decision-Making Units means organizational units that have a similar structure, perform similar functions, and, above all, are subject to the same procedures and regulations and have the same organization.
The DEA-CCR model is one of the basic DEA models and is based on the assumption of Constant Returns to Scale (CRS), meaning that a change in the number of inputs proportionally changes the number of outputs. This type of analysis makes it possible to assess the relative efficiency of decision-making units by determining an efficiency frontier on which the most efficient units are located. Units below this frontier are considered less efficient. The method also identifies the degree of inefficiency and suggests opportunities for improvement by reducing inputs or increasing results while resources remain unchanged.
DEA-CCR is particularly useful in situations where it is difficult to compare different units using traditional efficiency indicators. It makes it possible to assess efficiency under conditions of multidimensional input and output data. The method was developed to meet the needs of organizations that had to compare the efficiency of their units objectively while also taking account of the complexity of their operations. Its primary beneficiaries were managers, analysts, and decision-makers, who could use it to make better-informed decisions concerning resource allocation and improvements in operational performance.
Well-Selected Inputs Are the Key
The DEA-CCR assessment matrix transforms a vector of expenditures (inputs) into a vector of effects (outputs—results). Its purpose is to find the most effective configuration of effects produced on the basis of the inputs.
Let us return to our example of 120 border posts. Inputs for such facilities may vary. They may include the cost of cleaning services, telephone bills, and the cost of consumed fuel, electricity, and gas. Inputs may also include the total number of hours worked by all soldiers on duty during a month or the number of kilometers traveled by vehicles belonging to the post. The following factors appear to be the most representative inputs:
- Number of personnel—Border Guard officers and civilian employees working at the post. This is the key human resource affecting its operational efficiency.
- Operating budget—the post’s total annual operating expenditure. Including operating costs makes it possible to assess how well the base manages its financial resources.
- Area of the post—the square meters occupied by its buildings. This may reflect its capacity to carry out various operations, although a larger area usually entails higher maintenance costs.
- Infrastructure and maintenance costs—costs associated with maintaining infrastructure, buildings, roads, and equipment. They show how the base allocates resources to infrastructure and maintenance.
- Quantity and value of equipment—the number of vehicles and other military equipment. Equipment is a key element that enables the base to carry out missions while also constituting a significant maintenance cost.
- Energy resources—the amount of energy consumed or expenditure on energy. Efficient energy management is important for operations, particularly in the context of reducing costs and increasing operational efficiency.
In fact, inputs may include everything associated with the effort required to keep a base operationally ready. Matching the appropriate inputs depends on the available data and the operational characteristics of each base. Selecting suitable efficiency indicators (results) will likewise be crucial if the analysis is to provide valuable information.
Effects of the Border Post’s Work
Its function is to ensure that the border is secure. Effects may therefore include the number of people caught attempting to cross the border, hours of patrols conducted, reports prepared, and video materials produced. It is important to find those factors over which officers have influence.
When analyzing the efficiency of border posts and military bases using DEA-CCR, it is crucial to assess their actual operational effectiveness and select the appropriate effects (outputs—results). The effects should reflect the most important results achieved by the bases and should be linked to their mission and operational tasks. The following are several examples of potential outputs that can be taken into account:
- Number of completed military operations—the number of missions or operations, such as training, defensive, or humanitarian operations, that the base successfully completed during a given period.
- Operational readiness—an indicator defining the degree to which the base is ready to perform tasks at short notice. It may include the capacity to deploy forces rapidly, response time, and equipment readiness.
- Number of trained soldiers—the number of personnel trained at the base, which may be an indicator of its efficiency and ability to prepare personnel for various missions.
- Operational efficiency of equipment—an indicator measuring the technical condition and operational effectiveness of key military equipment at the base, such as aircraft, vehicles, and radar systems.
- Level of logistical support—the capacity to supply and maintain stocks such as ammunition, fuel, and food, as well as medical support. A high level of support may indicate effective management.
- Number of civilian-support missions—for bases carrying out missions supporting local communities, such as humanitarian aid or disaster protection, the number of such missions may serve as a measure of efficiency.
- Unit cost of operational activities—the cost efficiency of military activities carried out by the base; for example, the cost of a mission per unit of effect achieved (effectiveness relative to cost).
- Security level—the number of security incidents, including breaches and threats, or their absence. The fewer there are, the higher the base’s operational effectiveness.
- Logistical and technical support for external units—for example, the number of services provided to other units, such as supplies, servicing, or joint exercises.
- Training efficiency—for example, the number of training sessions or exercises conducted in which the base participated, and their impact on military effectiveness.
- Response time to alerts—the speed of the base’s response to various types of threats, which may be an important measure of its operational readiness.
The selection of appropriate measures depends on the specific activity of the base, its objectives, and the role it plays in the defense system. These effects should reflect the key results relevant to comparing its efficiency with that of other units.
How to Verify Input and Effect Vectors
In DEA-CCR analysis, it is easy to confuse inputs with effects. Is the number of patrols an input or an effect? Is the number of kilometers traveled by vehicles an input or an effect? The real difficulty in preparing this analysis lies in selecting the correct inputs and effects that will reflect the true efficiency of the border posts’ work. There are several methods for selecting appropriate inputs and outputs in DEA analysis. A situation in which a person ultimately selects the variables is generally ruled out. Ensuring that the analysis is objective requires variable selection to be supported by various mathematical and statistical methods. Some of these methods are presented below:
- Principal Component Analysis (PCA). It can be used to reduce a large number of input and effect variables. This method makes it possible to identify and combine into one factor the input variables that have the greatest influence on unit efficiency.
- Correlation analysis. Correlations between inputs and outputs can help identify variables that are significantly associated with results. The selected inputs should have a logical and statistically significant relationship with the outputs, increasing the reliability of the analysis.
- Regression analysis. Conducting regression between potential inputs and outputs can help determine which inputs have a significant impact on efficiency. Caution must be exercised in interpreting the results, however, because linear regression does not always reflect reality in the context of DEA.
- Expert methods (expert judgment). Although this is not strictly a mathematical method, expert-assessment methods supported by statistical and mathematical tools are frequently used.
- Stepwise selection. This variable-selection technique consists of iteratively adding or removing inputs in order to find the best set of variables that minimizes errors and improves the fit to the data. It is based on regression models to which variables are added or from which they are removed.
- Pareto criteria and efficiency analysis. In some cases, inputs and outputs may be selected on the basis of the Pareto criterion—that is, in a manner that maximizes the operational efficiency of the units.
- DEA-based optimization. So-called multistage DEA can be used, in which the initial stages help identify key variables and the main analysis is then performed using the selected variables.
Each of these methods has its advantages and limitations, and the selection of inputs and outputs should therefore be based on a logical analysis of the problem, data availability, and the context of the analysis. It is important for these methods to be applied in a justified and transparent manner in order to ensure the reliability of the DEA-CCR analysis results.
DEA-CCR Results
This analysis makes it possible to assess how efficiently border posts operate in comparison with one another. Its result is an efficiency list for every post. The posts that make the best use of their resources—such as personnel, equipment, and patrol time—to achieve effects—such as the number of interceptions, checks, or patrols—receive an indicator of 1 (100%) and form the so-called peer group. These posts serve as a model for the others.
If posts have a result below 1—for example, 0.8 (80%)—this means that their efficiency is lower. It indicates that they must improve their performance by 20% to match the best.
How Can Weaknesses Be Identified?
Thanks to DEA-CCR, every post with a score below 1 is compared with the most similar post from the efficient peer group. This means that a weaker post receives as its model a reference post selected specifically for it—one operating under similar conditions but making better use of its resources. For example, if a post conducts too few patrols with a similar number of officers, it can adjust its work organization by examining the activities of the reference post. The DEA-CCR algorithm will, however, select as the model a post operating under similar terrain and threat conditions.
How Can Results Be Improved?
Identifying weaknesses helps determine how efficiency can be improved:
- Improving inputs: resource use can be optimized—for example, by improving personnel management, using vehicles better, or optimizing response time.
- Increasing effects: with the same inputs, the focus can be placed on achieving more patrols, interceptions, or border checks.
For example, if a weaker post identifies that the reference post organizes its patrol schedule better, it can analyze and adapt those practices. This makes the improvement targeted and realistic. Moreover, the existence of model facilities for every post that is not fully optimal greatly increases the credibility of this method in the eyes of personnel.
It should be remembered that DEA-CCR is a tool that not only assesses efficiency but also points weaker posts toward specific solutions and models. This helps them improve their operations in an understandable and practical manner.
Wojciech Moszczyński
The author specializes in optimizing production and logistics processes.