Thus, productivity increases due to changes in technology or the sum total of operational and managerial efficiency (total factor productivity approach) are needed to be integrated while doing manpower forecasts. However, developing manpower planning models integrating the above two productivity variables is not so easy for obvious computational difficulty.
(ii) It is difficult to get units of output in the same form for all jobs. To take an example, maintenance jobs are difficult to quantify. Moreover, maintenance problem is time independent. Planning for maintenance staff is one of the difficult tasks for manpower planner. Due to this reason, most of the organizations prefer to give the contract of their maintenance function to another company or organization. Similar quantification problem is there for Customer Relationship Management (CRM). Even with CRM solutions, quantification is difficult.
(iii) The relationship between output and manpower is not always straight forward. Increase in output may lead to the economies of scale and the resultant cost efficiency and rise in productivity, which may not be attributable to the manpower productivity.
(iv) The effects of factors like new technology, incentive schemes, etc., upon productivity, may not be consistent over a time period. Therefore, projecting manpower requirement, considering effects of such factors may be inaccurate.
(v) The effects of different factors may not always be linear. The interrelationship of different factors complicates the forecasting of manpower. Although we have statistical techniques like multiple regression analysis, factor analysis, etc., their computational rigours often dissuade the manpower planners from using such tools.
(vi) Uncertainty about the future is again a major problem for the manpower planner. Thus, extrapolating on the past data may lead to a major inaccuracy in manpower estimation.
(vii) Data on past workload factors may not be available, creating difficulty in emulating the same.
(viii) Integration of manpower planning with corporate plans may not exist in an organization, creating problems for enterprise-wide manpower plans.
(ix) Employees cannot always be related to output in a direct way.
(x) Human Resource Information Systems (HRIS) may not exist in an organization. Lack of such information support system leads to inaccurate estimation of manpower.
For the obvious constraints explained above, managerial judgement along with the statistical tools, is often considered as the best option for manpower planning. Often it is said that planning for manpower within a given cost constraint, without bothering for past practices, lead to a better estimation.