Freezing the master production schedule for material requirements planning systems under demand uncertainty

Abstract Freezing the master production schedule (MPS) is one of the commonly used methods to reduce schedule instability or nervousness in material requirements planning (MRP) systems. However, the impact of MPS freezing parameters in multilevel MRP systems has not been fully understood. This paper investigates the impact of MPS freezing parameters on the total cost, schedule instability and service levels in multilevel MRP systems under demand uncertainty conditions. The MPS freezing parameters included in this study are: (1) planning horizons; (2) freezing proportions; (3) freezing methods; and (4) replanning periodicity. The paper also examines the impact of forecasting errors on the selection of MPS freezing parameters and system performance. A model was built to simulate forecasting, master production scheduling and material requirements planning operations in a manufacturing company. The company is assumed to produce two end items with no component commonality. The company operates in a make-to-stock environment. No back orders are allowed and thus any shortage will become a loss of sales. Through a comprehensive simulation experiment and statistical analysis, the study arrived at the following findings: (1) Forecasting errors significantly increase total costs and schedule instability, and reduce the service level in multilevel MRP systems. The selection of forecasting models has a significant impact on system performance. In addition, forecasting errors also influence the selection of some MPS freezing parameters. (2) A planning horizon of four natural ordering cycles results in a lower total cost and schedule instability and a higher service level than a planning horizon of eight natural ordering cycles under demand uncertainty condition, but a higher total cost and schedule instability under deterministic demand conditions. Thus the prolongation of the planning horizon can actually worsen MRP performance under demand uncertainty conditions while it improves MRP performance under deterministic demand conditions. (3) A higher freezing proportion results in a lower total cost and schedule instability, and does not influence the service level under deterministic demand conditions. Therefore, freezing the entire planning horizon is the best option in this case. However, a higher freezing proportion often results in a higher cost, and a lower service level, while reducing schedule instability under demand uncertainty conditions. As a result, the choice of freezing proportion should be based on the required trade-offs among the three performance criteria under demand uncertainty conditions. (4) The order-based freezing method results in lower total costs than the period-based method in most cases. In terms of schedule instability, the period-based method with freezing proportion of 1.0 performs the best. In terms of service level, the period-based method with freezing proportion of 1.0 performs the worst; the period-based method with freezing proportion of 0.25 performs the best. The selection of the freezing method should be considered in combination with the freezing proportion to achieve the required trade-offs among the performance criteria of total costs, schedule instability, and service level. (5) A higher replanning periodicity results in a lower total cost, schedule instability, and a higher service level under both deterministic and stochastic demand conditions. Thus, less frequent replanning improves system performance. These conclusions may help MRP practitioners to select MPS freezing parameters in multilevel MRP systems.

[1]  Douglas J. Dalrymple,et al.  SELECTING PARAMETERS FOR SHORT-TERM FORECASTING TECHNIQUES , 1981 .

[2]  R. C. Carlson,et al.  Less Nervous MRP Systems: A Dynamic Economic Lot-Sizing Approach , 1979 .

[3]  Urban Wemmerlöv,et al.  Lot-sizing under uncertainty in a rolling schedule environment , 1984 .

[4]  Dean H. Kropp,et al.  HEURISTIC LOT‐SIZING APPROACHES FOR DEALING WITH MRP SYSTEM NERVOUSNESS , 1983 .

[5]  T. S. Lee,et al.  AN EVALUATION OF FORECAST ERROR IN MASTER PRODUCTION SCHEDULING FOR MATERIAL REQUIREMENTS PLANNING SYSTEMS. , 1987 .

[6]  Kenneth R. Baker,et al.  AN EXPERIMENTAL STUDY OF THE EFFECTIVENESS OF ROLLING SCHEDULES IN PRODUCTION PLANNING , 1977 .

[7]  Dean H. Kropp,et al.  A Lot-Sizing Algorithm for Reducing Nervousness in MRP Systems , 1984 .

[8]  Everett E. Adam,et al.  Forecasting error evaluation in material requirements planning (MRP) production-inventory systems , 1986 .

[9]  Joseph D. Blackburn,et al.  Improved heuristics for multistage requirements planning systems , 1982 .

[10]  Dean H. Kropp,et al.  A comparison of strategies to dampen nervousness in MRP systems , 1986 .

[11]  Joseph D. Blackburn,et al.  HEURISTIC LOT‐SIZING PERFORMANCE IN A ROLLING‐SCHEDULE ENVIRONMENT* , 1980 .

[12]  Joseph D. Blackburn,et al.  The impact of a rolling schedule in a multi-level MRP system , 1982 .

[13]  Steven C. Wheelwright,et al.  Forecasting methods and applications. , 1979 .

[14]  Urban Wemmerlöv,et al.  The behavior of lot-sizing procedures in the presence of forecast errors , 1989 .

[15]  Dean H. Kropp,et al.  THE EFFECTIVENESS OF EXTENDING THE HORIZON IN ROLLING PRODUCTION SCHEDULING , 1982 .

[16]  Dean H. Kropp,et al.  MRP system nervousness: Causes and cures , 1985 .

[17]  V. Sridharan,et al.  Freezing the Master Production Schedule Under Demand Uncertainty , 1990 .

[18]  Neng-Pai Lin,et al.  A Model for Master Production Scheduling in Uncertain Environments , 1992 .

[19]  William L. Berry,et al.  MEASURING MASTER PRODUCTION SCHEDULE STABILITY UNDER ROLLING PLANNING HORIZONS , 1988 .

[20]  V. Sridharan,et al.  Freezing the Master Production Schedule Under Rolling Planning Horizons , 1987 .

[21]  Joseph R. Biggs,et al.  THE EFFECT AND COST OF FORECAST ERROR BIAS FOR MULTIPLE‐STAGE PRODUCTION‐INVENTORY SYSTEMS , 1982 .