Fred Zapata manager of industrial engineering at Eastern Tool & Stamping Co. in Saugus, Mass., has a job that sometimes looks like it belongs in Charlie Chaplin's movie Modern Times. He literally stands with a stopwatch and measures the time Eastern Tool's production workers take to do each job.

But the approach may be as modern today as when it was created. Jack Niles, of the Lexington, Mass., office of the consulting firm Rath & Strong Inc., argues that Zapata's classical way of establishing work standards could be a key to meeting the increased productivity pressures of the 1980s. Niles acknowledges that such old-fashioned work-measurement programs -- which calculate how much effort a task requires and how long it should take to do -- can be much more difficult to implement than they look. But, he says, they typically increase productivity 20% or more in companies that employ them. And managers normally can increase output an additional 25% to 30% by coupling work measurement to a formal incentive plan, he says.

The techniques Zapata uses to measure work have existed since the late nineteenth century. But recently, industrial engineers have been more interested in glamorous computer programs than in traditional work measurement, which involves carefully studying thousands of operations. Moreover, some managers question whether jobs can be adequately measured, while employees often resent a program they feel may place unfair demands on them. Because of this, many companies and schools of industrial engineering ignored or deemphasized work measurement in the 1960s and '70s.

But Niles says consulting clients have become more interested in work measurement in the last few years, as concern for productivity has increased. "To talk about productivity meaningfully you have to have some kind of measure," he notes. Last year employers sent 1,550 of their people to learn work measurement at a Charlotte, N.C., school sponsored by the consulting firm H. B. Maynard & Co., a leader in the field. As recently as 1979 the school trained only 950 employees.

As proof of the value of work measurement and incentive programs, Niles points to the system at Eastern Tool & Stamping, where Zapata's stopwatch sets norms for 100 of the 135 employees. Zapata determines an efficient method for a job, then times several workers while they do the task. He rates their work on a scale on which 100% is the work a good employee would do if he were measured but not offered incentive pay, and 130% is the work a typical worker can accomplish in response to incentives. The results, called pace ratings, include an element of subjectivity, but are based on objective norms: To learn to rate workers, work-measurement specialists watch films of employees performing tasks and then rate each performance. Then they compare their numbers to those compiled by more experienced time-study people. The idea is to come as close as possible to the established ratings.

Many work-measurement systems attempt to reduce the subjectivity still further by asking the person who is analyzing the job merely to list and classify the motions involved. The systems assume that each type of motion consumes a particular amount of time when performed by the average worker -- a moderately difficult reaching motion may require 0.010 minute, for instance. Thus a standard time for a job can be obtained by adding up the amounts of time required for each motion involved. But Niles of Rath & Strong points out that analysis that employs these systems still involves subjectivity in the classification of the motions.

Eastern Tool has an incentive system to go along with its work measurement. After Zapata rates workers he multiplies the average time they take for the task by the pace rating to get the time the task would theoretically consume for a measured worker uninfluenced by incentive pay. The company's base pay is in the $6-an-hour range. With incentive bonuses, a typical worker can earn nearly $8 an hour and an exceptional worker more than $9.

Both management and union say that Zapata measures jobs fairly and the system makes the company more efficient and a better place to work. "The men make more money," says Andrew Chiarelli, a former president of the International Union of Electrical, Radio and Machine Workers local at the company. Eastern Tool probably couldn't afford to pay $8 an hour without the extra productivity that the incentive system creates, he says. "What I do in 5 minutes would probably take 20 minutes to do if a guy were getting paid by the hour," Chiarelli asserts.

Management and employees did not always have such a high opinion of work measurement at Eastern Tool. When the system was introduced in 1966 the company suffered the difficulties typical of a new work-measurement program. Job evaluation was done by ordinary employees the company had trained. "There was a dictatorial attitude," notes Gene Decareau, plant manager. "You know, 'I set the goddamn standard, now do it.' "

Says Zapata: "Few people who learn time-study realize the depth of human relationships needed for good work measurement. You have to show people they can achieve the 130% of standard."

Also, an inept time-study person can easily be deceived by workers into setting standards too low. The typical starting salary for a good work-measurement industrial engineer, say Niles and Zapata, runs between $25,000 and $35,000.

Zapata knows he has earned everyone's trust. As he walks around the shop floor, an automatic stamping machine operator, Dionicio Echevarria, complains about the job he has been assigned. "Hey, Joe, this job's no good," he declares. "You don't make money." Zapata investigates and discovers that the spced of the machine Echevarria is operating had been turned back from 200 to 140 units per minute because of a problem with the die, but the production standard had not been changed. He recalculates a temporary standard that makes it easier for Echevarria to earn an incentive bonus.

Unfortunately it is easier to measure the work of press operators than of most workers. Thomas Slaight, a principal and partner at the Los Angeles-based consulting firm of Theodore Barry & Associates, notes that managers have to be "much more creative" in developing and applying productivity measures for white-collar workers and people with diverse tasks. But some rough productivity measure can usually be produced. White-collar workers, for example, can be asked to keep diaries of their work. From that information, one can calculate the amount of time required to type an invoice or a letter, or the number of tasks an employee performs away from his or her desk.

But work-measurement and incentive systems inevitably produce problems. Even where work is as straightforward and the management as careful as at Eastern Tool, an incentive system seems to discourage employees from offering ideas for productivity improvement once a standard has been set. Chiarelli, the former union president at Eastern Tool, notes that many workers have developed easier ways of doing jobs than the methods Zapata prescribed. "The guy who actually does a job is always the one who knows it best," he says. But workers at Eastern Tool rarely point out these improvements to management.

Slaight argues that incentive systems like Eastern Tool's are "dragging back productivity in many cases" because the difficulties of recalculating the standards make changes in methods harder. Instead of paying large bonuses to people who exceed norms, he suggests small rewards such as certificates of merit and prizes. Such encouragement, he notes, tends to increase, rather than reduce, employees' motivation to offer suggestions.

Such programs can often be administered on a group basis. At Metpath Inc., a medical-testing laboratory in Teterboro, N.J., for instance, Slaight divided specimen-receiving clerks into competing teams. The winning section was given a certificate of merit for several months, then the prize was changed and the winners received free pizza. Slaight suggests that work measurement should be handled in a way that makes exceeding norms a game for workers doing repetitive tasks, because people with boring jobs need games to help keep their interest. "If you don't help them develop constructive games, they'll develop destructive games," he says.

But Slaight argues that before employers innovate with creative work-measurement systems, they should look through their files to study work measurement previously done in the company. The most common failure in work measurement is not lack of creativity but a failure to follow through, he says. "Management goes through phases of promoting these things, and then they put them on the back burner," Slaight comments. Consulting firms can sometimes sell a workmeasurement system to a company, expect it to work for two or three years before declining, and then go back five years later after management has changed and sell a nearly identical system.

Revitalizing the remnants of an old measurement system may be easier than inventing a new one. "Acceptance comes two or three years down the road, when the employees and the supervisors realize, 'Hey, this time old George really means it,' " says Slaight.