Getting back to the original question, the solutions are complicated. The first thing to factor is cost versus benefit. How much value will be obtained by using a drone to deliver articles at a construction site? How much time is on average spent by personnel delivering small articles to drone accessible locations at a construction site? How often would the platform perform those duties? How many job sites will it be used for? What is the estimated useful life span of the drone? Those are important because they need to become more economical and efficient than using lower wage human labor, and general labor is pretty cheap even when factoring in indirect and over head labor expenses. We’ll also need to factor in the hourly wage of the drone operator, which would likely be higher than the hourly cost of unskilled labor.
Next you need to determine how much money can be expended in building such a drone? That cost forms the basis needed to amortize and balance the cost per flight against human labor. Using that very expensive $36,000.00 drone as a cost foundation we can see it would take a great many deliveries before any money could be saved instead of using a human.
After obtaining answers to the above we can establish the average size and weight of the articles we want to deliver. Alternatively we could use a maximum drone cost and find an existing drone that fits within the cost category and use that drones’ lift capacity to establish a maximum cargo capacity, but that limits your selections to what is currently available. You would still need to determine cost versus benefit to to learn if the concept is practical.
It will take an awful lot of 1lb. nail deliveries to offset the cost of even a $2,000.00 drone. If human labor costs $15.00/hour and indirect labors factors at 1.3 times the direct cost, and human labor averages 30 minutes/delivery, you’ll have to perform 99 one pound delivery flights before just the cost of a $2,000.00 drone is recovered. Add in the wage cost of the drone operator and that number is much, much greater.
Once you have the payload size and weight you can move on to an airframe? How big? If lifting only a pound or so a 500mm-600mm rig would be adequate. You won’t need a fancy video camera and its associated weight, an FPV camera to see here it’s going is more than adequate. If the intent is to carry heavier cargo the frame has to be larger, starting somewhere around 700mm-800mm. Add the average weight of a flight controller to the weight of a frame, FPV camera/video transmitter, and cargo weight. If you want to jump forward a bit you might add the weight of a couple 6S, 8000mA batteries to that total as it’s likely you’ll end up going with a 20V-22V system.
Now you can peruse motor and propeller manufacturer performance data sheets to learn which motor and prop combinations provide the thrust needed to fly the machine. You know you’ll need at least 4 motors (6 would be better if safety is a concern) and the total motor count needs to provide, at minimum, double the amount of thrust needed to lift the weight. Don’t forget to add the weight of the motors, props, ESC’s, and wiring (12 ounces for wiring is a good starting point) to the total system weight, which will generate a more than a few headaches in power necessary for weight calculation revisions. e-Calc becomes very helpful at this point as it can estimate system efficiency and flight time using you component selections. It’s not absolutely accurate but it provides enough accuracy to determine if your design selections can function well enough to meet general performance parameters.