Procedures for Creating Heat Maps for the Ministry of Land, Infrastructure, Transport and Tourism｜7 Steps Even Beginners Can Understand

The job of creating heat maps that comply with the Ministry of Land, Infrastructure, Transport and Tourism (MLIT) does not end with simply outputting a colored distribution map. In practice, you must determine which construction categories the procedures apply to, how to align 3D design data and as-built measurement data, what range to evaluate, and how to format the deliverables as inspection-ready reports before you can truly say the heat map has been “created.” Because MLIT’s procedures are organized by construction type and the as-built management manuals are continually revised, the starting point for any task is always “confirm the target construction and the latest manuals.”

Table of contents

• What a heat map is

• Initial premises to grasp for MLIT compliance

• Step 1 Confirm the applicable construction type and management targets

• Step 2 Prepare the 3D design data

• Step 3 Decide the measurement plan

• Step 4 Acquire as-built data on site

• Step 5 Create data for as-built evaluation

• Step 6 Convert differences into a heat map

• Step 7 Tabulate and finalize before inspection

• Common pitfalls for beginners

• Summary

What a heat map is

The heat map used in MLIT as-built management is an as-built management chart that overlays 3D design data with as-built evaluation data and shows the deviations at each point as a distribution map. In practice, what matters is not visual flashiness but the ability to instantly understand where each point stands relative to the standard values. In explanatory materials from the Kanto Regional Development Bureau, as-built management charts are presented as distribution maps that plot elevation or horizontal deviations of each point on a plane, and the idea is organized as confirming a heat map that colors the deviation calculation results by percentage relative to the standard values.

In other words, a heat map is not a document for appearance; it is a tool to verify as-built conformity across an area. Its major value is enabling a planar understanding of undulations and biases that were hard to see when managing only limited traditional measurement points. For practitioners, it is important not only to know pass/fail but quickly to grasp where the surface tends to be overfilled, where over-excavation or insufficient construction tends to occur. Therefore, creating a heat map should be understood not as “coloring” but as “correctly producing a distribution of differences usable for evaluation.”

Initial premises to grasp for MLIT compliance

The first thing to grasp is that MLIT’s as-built management manuals and supervision/inspection manuals are organized by construction type, and the premises and checkpoints differ for earthwork, pavement, slope works, structures, etc. On MLIT’s manuals page, supervision and inspection procedures for as-built management are listed by construction type, and as-built management manuals are continually revised. Therefore, thinking “we did it this way on a different project before, so the same approach will work now” is risky. Start by confirming the target construction type and what will be evaluated for that type.

Also, the guidance positions the use of 3D measurement technology for as-built management as materials to confirm selection of measurement methods based on site conditions, measurement procedures, practical effects, and precautions. Beginners are less likely to fail if they understand the overall picture—why a measurement method is chosen, which parts are evaluated as surfaces, and which parts require different management—rather than jumping straight into software operations.

Furthermore, the success of heat map creation is not decided solely by post-processing office work. The quality of preceding steps—on-site measurement planning, handling of extraneous points, design data consistency, and the concept of excluded evaluation ranges—directly affects the result. Even a neatly formatted report will not become a reliable heat map if the source data are misaligned. Understanding this premise helps avoid doing steps out of order.

Step 1 Confirm the applicable construction type and management targets

The first step is to determine which construction type the project falls under and which manuals will govern management. MLIT does not treat everything uniformly in a single document; rather, the as-built management concepts and supervision/inspection confirmation methods are organized by construction type. If this remains vague and you proceed, the heat map you produce may turn out to be “not the management document required for that project.”

For example, areas suitable for surface management are not the same as areas where conventional dimensional control should be prioritized. MLIT’s Q&A shows that for some construction types, management may be dimensional control using point clouds measured on surfaces rather than heat map evaluation using 3D design data. In short, heat maps are not universal and must be used selectively according to construction type and part. Beginners should first inspect the target parts and classify whether “this is subject to heat map evaluation” or “different management is required.”

At this stage, also confirm the start and end points of the control cross-sections, the surfaces to be evaluated, and boundaries of parts with different standard values, as this will simplify later steps. MLIT’s guidance suggests creating as-built management charts by part where standard values differ. Even if a surface looks continuous on site, it may be separate parts for management; mixing them into one heat map makes evaluation ambiguous. Pre-classifying parts is ultimately the shortest route.

Step 2 Prepare the 3D design data

The next step is to prepare the 3D design data that will serve as the evaluation benchmark. Because heat maps are created from differences between the design surface and the as-built evaluation data, incomplete design data prevents correct evaluation no matter how accurately you measure. As a rule of thumb, it’s reasonable to doubt the quality of the design data before trusting the measurement data. Carefully check for missing information relevant to evaluation such as alignment, berms, toe of slope, top surface, flat areas, slopes, and edge shapes.

Beginners often overlook that design data may appear visually correct but be difficult to use as a reference surface. For example, if the changeover of standard values is ambiguous, the control range is unclear, or the treatment of edges and change points is not organized, the color distribution after difference calculation will look unnatural. Even if you interpret that oddity as a measurement mistake and re-measure, the cause is often the design data.

Kanto Regional Development Bureau materials indicate that the measurement range should be from the start to the end points of the control cross-section described in the 3D design data, clarifying the management range before acquiring as-built coordinate values. This shows that the essence of heat map creation is “what to use as a reference, where, and how to compare.” Clarifying the design-side management range before measurement is the most unglamorous but most effective preparation.

Step 3 Decide the measurement plan

The third step is to decide how to measure on site. What matters here is not aiming to collect a dense point cloud for its own sake. The Shikoku Regional Development Bureau’s operational guide explicitly states that the goal is improving quality control and streamlining work, and obtaining an extremely dense point cloud is not the objective. What is required is a plan to acquire data sufficient for evaluation under feasible, reproducible conditions.

To do this, decide in advance where to measure from, whether blind spots will occur, whether temporary materials, heavy equipment, or worker movement will interfere, how to check backsights and references, and the timing of measurement. In particular, for surface-managed heat maps, failures due to local missing data are less critical than omissions or poor stitching across the entire surface. Some surfaces become unmeasurable once construction progresses, so consider whether the surfaces you want to evaluate are actually observable at the intended time given site conditions.

In practice, being aware of ranges that may be excluded from evaluation improves efficiency. Earthwork materials show excluding certain ranges near change points such as berms and toes of slopes, and structural work guides also organize excluding fixed end areas from evaluation. Beginners tend to think “using all points is more rigorous,” but forcing in ranges unsuitable for evaluation actually reduces the overall reliability of the heat map.

Step 4 Acquire as-built data on site

The fourth step is to acquire as-built data on site according to the plan. Keep in mind acquiring data in a way that makes subsequent processing easier. Capture surfaces from positions where they are sufficiently visible, minimize intrusion of extraneous objects, and if multiple measurements are needed, create conditions that make later stitching easy. On site people often think “we can delete later,” but the more extraneous points there are, the longer processing takes and the more causes of evaluation disturbance arise.

Shikoku Regional Development Bureau’s guide notes that poor stitching of measurement data prevents accurate as-built evaluation, so confirm backsight points and the stitching status of point cloud data during measurement. This very practical point shows that heat map failures often stem from earlier alignment and stitching issues rather than the difference calculation. Developing a habit of checking for shifts, missing data, blind spots, and extraneous objects immediately after measurement greatly reduces the need for rework.

Also, avoid trying to be perfect in a single pass unnecessarily. Depending on site conditions, acquiring data by subdividing parts can produce more stable results. Because heat maps are ultimately evaluated by part, keeping reproducibility at the part level during acquisition makes later editing and report generation easier.

Step 5 Create data for as-built evaluation

The fifth step is to create as-built evaluation data from the acquired data. This is the part beginners most easily misunderstand. You cannot simply feed the raw measurement data into a difference calculation. You must remove extraneous point clouds, retain only the structural parts to be evaluated, and refine the data into evaluation data that meets required density and distribution. Shikoku Regional Development Bureau’s guide shows the workflow of deleting extraneous point clouds such as temporary materials to leave only structural point clouds and then extracting data for difference calculation and heat map creation.

Kanto Regional Development Bureau materials state that as-built management documents are created using 3D design data and as-built evaluation data, and that evaluation data are used on the premise of judging conformity by the deviation at each point. In other words, evaluation data are not just “what was measured,” but “what was prepared into an evaluable form.” Skipping this step leads to wildly scattered differences, unnaturally intense coloring in particular areas, or results that do not match reality.

Earthwork materials propose obtaining at least one as-built coordinate per 10 cm (3.9 in) mesh across the entire range, and organizing as-built evaluation data as at least one point per 1 m² (10.8 ft²). However, these specifics depend on construction type and the relevant guidance, so in practice you must confirm against the latest manual for your target construction. The important thing is not to be satisfied with the amount of raw data but to ensure the density and quality match the evaluation conditions.

Step 6 Convert differences into a heat map

The sixth step is the actual difference calculation to create the heat map. Here you overlay the 3D design data and as-built evaluation data, calculate the deviation at each point, and turn it into a distribution map. Structural works guides introduce the workflow of calculating differences to create a heat map and note various methods such as averaging, nearest neighbor, TIN, and inverse distance weighting. In practice you may be tempted to follow software defaults, but if you don’t understand which method you used to calculate differences, you will struggle when comparing results with others or explaining them at inspections.

Color assignment settings are also important. Kanto Regional Development Bureau materials recommend coloring calculation results as percentages relative to the standard values in a range from −100 percent to +100 percent, making the legend explicit, and using distinct colors for around ±50 percent and around ±80 percent, with out-of-spec ranges shown in a separate color. The point here is not merely to color red and blue but to express the relationship to the standard values in a readable way.

From a practitioner’s perspective, a heat map that simply shows “there are differences” is insufficient. It must indicate which differences are within tolerance, which warrant caution, and which are out of specification. Therefore, rather than using too many colors that confuse readers, set a legend and thresholds that allow inspectors to judge at a glance. Evaluation documents are not just for your own understanding; they must be interpretable the same way by third parties.

Also, when difference results feel off, don’t immediately tweak only the recalculation conditions; go back and check whether the cause lies in design data, coordinate systems, evaluation range, extraneous point removal, or stitching status. Heat map creation may look like a final step, but in reality it reflects the quality of previous steps. If you see unnatural stripes or biases, the problem is often input conditions, not color settings.

Step 7 Tabulate and finalize before inspection

The seventh step is to organize the heat map into as-built management documents that can be read during inspection. Kanto Regional Development Bureau materials define as-built management documents as as-built management charts and recommend organizing average, maximum, minimum, data count, evaluation area, number of rejected points, and then presenting the deviations of each point as a distribution map. In other words, a heat map image alone is insufficient; only when numerical summaries and distribution maps are combined does it become a management document.

Delivery formats such as PDF or 3D data with a viewer are also suggested. On site there is temptation to get by with screen captures, but what must be checkable before inspection is whether the legend is readable, whether parts with different standard values are separated, whether the reasoning for excluded evaluation areas is explainable, and whether the treatment of outliers is organized. Rather than merely outputting what the software produces automatically, reports must be refined so supervisory and inspection personnel can read them without confusion.

Keep in mind that submitted materials are “explanatory documents,” not merely “work results.” For example, structure the report so that it conveys—without oral explanation—why parts were separated, why some ranges were excluded from evaluation, and why a particular color scheme was chosen; this greatly reduces pre-inspection anxiety. Beginners tend to postpone report preparation, but a heat map becomes a practical document only when tabulated.

Common pitfalls for beginners

The mistake beginners most often make is thinking heat map creation is mainly a software operation issue. In fact, many failures occur earlier. Causes include not confirming the applicable construction type, unclear design data ranges, blind spots or missing data in site measurements, insufficient removal of extraneous points, and forcibly including ranges that should be excluded. These problems are not solved by changing color settings.

Another common misconception is assuming more raw data is always better. While necessary density is important, having a large number of unnecessary points or mixing structural and temporary parts destabilizes results. MLIT operational guides emphasize deleting extraneous point clouds and checking stitching status. Heat map quality is determined not by data quantity but by how well the evaluation target has been extracted.

Ignoring differences between construction types is also a typical failure. Applying earthwork procedures unchanged to structures, or partially reusing another construction type’s guidance, may allow work to proceed but make inspection explanations impossible. The reason MLIT organizes documents by construction type is that the approaches differ significantly. When in doubt, returning to the latest manual and confirming the target construction type is ultimately the shortest path.

Finally, be careful not to make heat map creation an end in itself. The real purpose is to appropriately grasp as-built conditions and link that to quality assurance and efficient construction management. Spending too much time polishing reports so on-site decisions are delayed, or beautifying documents for inspection while not using them for construction judgment, is counterproductive. Heat maps are valuable only when used on site.

Summary

For beginners to understand MLIT-compliant heat map creation, the important thing is to consider the seven steps in order. First confirm the applicable construction type and management targets, prepare 3D design data, plan measurements, acquire necessary as-built data on site, refine the data for evaluation, perform difference calculations, and finally produce inspection-readable reports. Following this flow makes heat maps easier to understand as a practical tool for assessing as-built conformity across areas rather than a difficult specialist document. Because manuals are continually revised, always proceed while confirming the latest manual for the target construction type.

On site, overall work efficiency depends greatly not only on the heat map production steps but on how quickly and accurately you can carry out the preceding coordinate checks, stakeout, and reference point confirmations. When you want to streamline these routine site tasks, adopting options such as LRTK—an iPhone-mounted GNSS high-precision positioning device—can facilitate initial establishment surveys, on-site coordinate checks, and simplified surveying. While final heat map creation must be performed carefully in accordance with construction type and manuals, LRTK can be one option to help seamlessly connect on-site tasks before and after heat map creation.