One of the primary sources of information used by adaptive intelligent traffic management systems is vehicle traffic count data from each lane of an intersection. The intelligent traffic control system makes poor conclusions about the best time allocation when detectors acquire data incorrectly or when traffic volume count data does not match the actual number. The properties that loop detectors extract are insufficient in intelligent traffic control systems to compute missing data with accuracy. The majority of inference techniques currently in use only make use of the features that have been retrieved, which results in data models that are not accurate enough. This shortcoming is the reason to propose an enrichment inference method for loop detectors, called EIM-LD, which introduces a data enrichment strategy employing statistical multi-class labeling to boost the inference accuracy for various missing patterns and rates.  The original data, including the missing traffic volume data, is first gated without the suggested enrichment process to assess the impact of applying the suggested method.  Next, the suggested method’s accuracy is assessed by taking into account both.