**5. Results and Discussion**

This study identified and evaluated the key data quality dimensions for assessing the data quality of highway infrastructure for decision-making effectiveness. For this purpose, the study considered the critical dimensions throughout the highway infrastructure project, as well as the criticality of dimensions at each level of decision-making, as the data quality requirement varies at each level of decision-making. Using the ranking of dimensions shown in Table 3, the overall critical dimensions and preference of the dimensions for the overall project data were determined. Table 4 illustrates the significance of dimensions at each level of decision-making. From the analysis of Table 4, it is clear that the requirements for decision-makers are no longer the same but vary according to the respective hierarchical levels of decision-making.

The data quality dimensions listed in Table 3 are relevant to ensure the overall quality of data for highway infrastructure projects. Effective decision-making relies on the availability of high-quality data, and addressing each of these dimensions can help to ensure that the data used in decision-making are accurate, complete, consistent, and timely. Based on the mean scores in Table 3 and the analysis of Figure 3, the top five data quality dimensions are accuracy, accessibility, completeness, consistency, and timeliness. Ensuring that data are accurate involves verifying that they are correct and error-free. Accessibility involves making the data available and easily retrievable to authorised stakeholders. Completeness ensures that all required data elements are present and accounted for. Consistency involves verifying that the data are consistent with other data elements within the project. Timeliness ensures that the data are available when needed and up to date. Other dimensions listed in Table 3, such as relevance, interpretability, and believability, are also crucial for effective decision-making. Relevant data are essential to the decision-making process because they ensure that the data are related to the project's objectives. Interpretability ensures that data are presented in a way that is easy to understand. At the same time, believability involves ensuring that the data can be trusted and are not biased.


**Table 4.** Decision-maker's level of importance of dimensions.

**Figure 3.** Semiotic levels and the data-information-knowledge hierarchy.

At the same time, the stakeholders consider conciseness, ease of operation, and security as the lowest priorities with low mean values. This might be due to the dimension conciseness that implies the compact representation of data, which would create a problem of understanding for all stakeholders for data usage in decision-making. The dimension ease-of-operation implies that data are manipulated and easily customised, which stakeholders feel could cause problems in decision-making if the data are easily manipulated. The dimension security implies keeping data secure and restricting access to the data, and the stakeholders feel the restricting of data would cause issues with the decision-making.

Figure 3 shows the box-and-whiskers plot of the data quality dimensions for assessing overall highway project data quality. It shows that the range of most of the dimensions is between 3 and 5, i.e., the responses from the decision-makers range from somehow important to high importance. Based on the data, it seems that accuracy, completeness, and accessibility are considered to be the most critical dimensions of data quality by more than half of the decision-makers, as they have the highest median value of 5. The dimensions consistency, structure, integrity, timeliness, ambiguity, relevant, and value also have a relatively high median value of 4, indicating that they are still considered essential by many decision-makers. On the other hand, dimensions such as conciseness, ease of operations, security, definition, believability, understandability, validity, and appropriateness have a median value of 3, indicating that they are considered less critical by decision-makers. It is important to note that these findings are based on decision-makers' responses and reflect the objective of data quality measures. Nonetheless, they provide valuable insights into the perceived importance of different dimensions of data quality in the context of highway project data.

## *5.1. Critical Dimensions at Each Level of Decision-Making Hierarchy*

The importance of data quality dimensions at each decision-making level, such as strategic, network, program, project selection, and project decision levels of highway infrastructure projects, was also identified, along with key data quality dimensions for assessing the over-project data. Table 4 shows the description of assessment measures and a survey result on the level of importance for semiotic framework data quality attributes obtained from highway decision-makers, respectively. Based on the level of importance, decision-makers think that all data quality dimensions described within the semiotic framework are crucial for generating information at all levels of the decision-making hierarchy for highway infrastructure. However, the results indicate that the conciseness, ease of operational ability, and data understandability dimensions do not significantly influence decision-making processes at all levels of highway infrastructure decision-making, i.e., strategic, network, program, project selection, and project. This may be due to the absence of a system that facilitates the understanding of collected data at these levels. The collected data could be in various formats, including text, images, or numbers, and they could be used as input for decision-making. For data to be used as input, they must be clearly understood according to the judgment of the highway engineers. This may result from the insignificant use of project data at these levels or the continuation of decisionmaking processes due to limited project scope in the early stages of a project. Therefore, the dimensions with a rating of 4 out of 5 are regarded as crucial for generating information at all levels of the decision-making hierarchy.

#### 5.1.1. Strategic Level

At the strategic level, decision-makers are higher-level authorities, such as the chairman and division heads of NHAI. They deal with policies, guidelines, and the distribution of funds. At the strategic level of the decision hierarchy, the accuracy, consistency, completeness, structure, and integrity dimensions from the syntactic category; the accessibility and timeliness dimensions from the empiric category; the definition, reliability, and ambiguity dimensions from the semantic category; and the relevance and value dimensions from the pragmatic category are crucial decision-making dimensions.
