SURVEYING ENGINEERING BLOG

BLOG ENTRIES
Mistakes & Errors
by: Rich Cebrick   12/2/2013

Mistakes and errors will never be totally eliminated from surveying measurements. However, with proper training and awareness, the instrument operator, note keeper and other field personnel will be able to greatly reduce the instances where errors, mistakes and blunders occur. Furthermore, these discrepancies will potentially be reduced as the field crew learns about the types of errors that occur and the sources of those errors.


It is important to realize that mistakes and errors are not the same thing. They have different sources, effects and magnitudes in computations. According to Dr. Ben Buckner, LS, PE, CP, “Mistakes in measurements can be traced to carelessness, inattention, improper training, bad habits, lack of innate ability, poor judgment, adverse measuring or observing conditions and various negative emotions, negative attitudes and perceptions that plague humans.” Mistakes often occur when an inadequate amount of information has been obtained and some evaluators even go as far as to make the slip-up of using their intuition to predict results. Fortunately, mistakes and blunders are often easily identified and resolved. However, there are times when a mistake is so miniscule that it goes undetected in computations. Errors, on the other hand, are unavoidable even when great care is taken in attaining measurements. Errors come in two types, systematic and random and have four sources; natural, instrument, personal and calculation. They arise in every measurement, due largely in part to human and instrument imperfections. Systematic errors obey the laws of physics and can be quantified and removed (i.e. instrument maladjustment or lack of calibration). Random errors obey the laws of probability and remain in measurements after the systematic errors are removed (i.e. cross-hair alignment on targets or centering of the instrument.) Accordingly, natural errors arise due to factors in the environment, instrument errors originate from instrument imperfections, personal errors are due to the fact that humans are flawed and calculation errors originate from computations (i.e. failing to use proper conversion factors or rounding digits in the middle of the computation process rather than at the end.)


It is very important that surveyors take special care when dealing with measurement data, always knowing the limitations of the equipment used along with the proper procedures to compensate for those restrictions. Furthermore, trained surveyors should have the experience to understand where mistakes, blunders and errors originate. Through appropriate schooling, they acquire the technical skills to avoid practices that cause such missteps and understand the application of proper techniques to remove errors when possible.


Citation:

Buckner, B. "The Nature of Measurement: Part 2, Mistakes and Errors", Professional Surveyor Magazine 17 (No. 2) Archive.

Counting vs Measuring
by: Rich Cebrick   12/1/2013

The exact value of any measurement is never truly known and there can, at times, be a significant difference in numbers representing counts and those representing actual measurements. For this reason there is a definite need for any measurement assessor, from surveyors to physicians to postal workers, to be able to identify errors, their sources and the amount of uncertainty in a quantity.


No matter how precise an instrument is or how much care is taken in the operation of that instrument, measurements made are only an approximation of the exact value of a quantity. This is due more than anything else to the fact that there are imperfections in humans and in instruments (though instrumental imperfections may be miniscule). Additionally, one must accept that, beyond a reasonable doubt, there will always be distractions in the environment in which the measurements are being made. In the world of Geomatics and Land Surveying, surveyors need to be able to assess the amount of uncertainty in measured quantities in order to utilize data correctly.


Taking care to realize the difference between numbers as counts and numbers as measurements may reduce some discrepancy in assessed values. Counts are represented by whole numbers with no estimating involved (in the rare cases where decimals are present in counts those fractions are based solely on the abilities of the evaluator conducting the count). For this reason, counts cannot be sustained beyond their last numeral. However, counts may be considered exact in terms of quantities of physical objects such as the amount of money in a jar or the number of marbles in a bag. Measurements, on the other hand, are never to be considered as exact. According to Dr. Ben Buckner, "In contrast to counts, measurements always involve numbers that are inexact. In theory, such numbers continue to infinity. A measurement expressed as 147.92 feet might be 147.92247 or 147.91832, or any other string of numbers that would round to 147.92. We do not know what lies behind that last digit expressed, and even the last digit could be off a little."


In essence, there must always be an underlying acceptance that all measurements will contain error. Accordingly, care must be taken to evaluate the sources of errors and contain or at least compensate for those errors.


Before participating in a data analysis class it was very easy to take most measurement values at "face value". However, after entering into the surveying curriculum it is quite apparent that all measurements must be approached with bit of skepticism. This article definitely supports that notion. Surveyors must take special care when dealing with measurement data and always know the limitations of the equipment used along with the proper procedures to compensate for those restrictions. Additionally, surveyors have an obligation to be aware of the sources of their data when that data was collected by other people and the limitations in the techniques conducted in collecting that information.


Citation:

Buckner, B. "The Nature of Measurement: Part 1, The Inexactness of Measurement – Counting vs. Measuring", Professional Surveyor Magazine 17 (No. 2) Archive.

The GPS Accuracy Game
by: Rich Cebrick   5/27/2013

Surveyors and data analysts alike are constantly looking for newer, faster, cheaper and more accurate measurement tools and, regardless of how worthy those instruments are of praise, each has its limitations in precision. According to Gavin Schrock, "GNSS, almost without argument as one of the most amazing technological advancements ever, is no exception to this dynamic."


GPS was first employed in 1978 and saw great improvements and wider use in 1982 with the introduction of commercial receivers. Later, in 1988, dual frequency receivers became available. In 1992 Real-Time Kinematics (RTK) used dual frequencies and differential methods (Differential GPS) to produce data that was comparable to fully-processed measurements. Differential GPS (DGPS) was acceptable for agriculture, marine navigation and limited mapping applications; however, it was nowhere near accurate enough for surveying applications without the implementation of RTK and additional post processing. Nonetheless, even after processing, most systems still did not possess the required surveying accuracy while maintaining a cost effective status. Soon after, in the late 1990s, Network RTK (RTN) was introduced. The most recent system, and possible RTK replacement, is known as Precise Point Positioning (PPP). In the near future it is believed that PPP will allow for precise measurements without the need for base/CORS setup, using only a rover communicating with a satellite network. This system would rely heavily on "error state products (ie clock, orbit, ionosphere and troposphere models)". PPP requires knowledge of a satellite position upon signal generation, time of signal generation and time of satellite signal reception by the rover. Great coordination of satellite position and time offset must be maintained with little or no error in the data.


Currently RTK/RTN provides good service with centimeter precision, fast initialization and a dense network, but, overall, the service is slow. PPP provides only decimeter precision (representing a step down from RTK/RTN) and that service is not cheap. PPP-RTK provides centimeter precision quickly, but the overall service quality is poor. Thus there is an obvious trade-off of one or more desirable properties for another.


When all is said and done, regardless of how advanced our current GPS technology has become, there still needs to be advancements in the systems if they are ever to be widely acceptable for the needs of today’s modern surveyor. Surveyor’s must always be aware of the limitations of their equipment, know where all errors in data come from and be equipped with the knowledge of how to correct and compensate for those limitations and errors.


Citation:

Schrock, G. "GNSS Part 1: After RTK", Professional Surveyor Magazine 32 (No. 6): 10-15.

2013-2014 Penn State Wilkes-Barre Surveying Society
Website created by Rich Cebrick Spring Semester 2013