Post on 30-Dec-2016
transcript
Productivity/performancemeasurement
Case application at the governmentpharmaceutical organization
Kongkiti Phusavat and Watcharapon PhotaranonDepartment of Industrial Engineering, Kasetsart University,
Bangkok, Thailand
Abstract
Purpose – To address two key problems facing the production department at the governmentpharmaceutical organization (GPO). They were a lack of productivity and performance measurementat the operational level, and a need to assess the functional readiness to undertake its ownperformance-analysis work.
Design/methodology/approach – The decision was made to apply the multi-factor productivitymeasurement model (MFPMM). This was due to its ability to link with existing data that had beengenerated by the accounting department, its explicit integration between productivity and financialperformance (i.e. profitability) at the operational level, and its major features such as information onopportunity gain/loss.
Findings – For the GPO, it was quite clear that productivity improvement would represent thecritical direction for the production department in the future. Based on its initial deployment at theSection 3 unit, the MFPMM was generally viewed as a suitable tool for measuring and providingnecessary productivity/performance information.
Practical implications – The database management needed to be improved in order to ensuretimely provision of data into the MFPMM. Furthermore, better performance-analysis capability by thestaffs was required. In other words, both technical and managerial aspects must be recognized in orderto ensure the MFPMM success and sustainability.
Originality/value – This study generated insightful information into operational performance at theGPO. It was clear that productivity represented the basis for formulating future policy initiatives.
Keywords Productivity rate, Prices, Profit, Measurement
Paper type Case study
BackgroundProductivity and subsequently performance measurement has become more importantand has been regarded as a prerequisite for continuous improvement (Kaydos, 1991).When focusing on the industrial, national, and international levels, many approacheshave been designed by economists such as the total factor productivity (TFP), orBureau of Labor Statistics (BLS) multifactor productivity techniques (Duke andTorres, 2005; Meyer and Harper, 2005; Tsai et al., 2006). At the organizational,functional, program, and project levels, there have been several concepts and ideasinvolving in the measurement/assessment work. The balanced scorecard concept(Kaplan and Norton, 2004) has often been cited. The Sink and Tuttle (1989) approachhas been widely adapted by public/private organizations. Harper (1984) also develops aperformance measurement framework at the organizational/functional levels. Theconcepts initiated by Sink (1985), and Harper (1984) share one similar aspect – the use
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/0263-5577.htm
IMDS106,9
1272
Industrial Management & DataSystemsVol. 106 No. 9, 2006pp. 1272-1287q Emerald Group Publishing Limited0263-5577DOI 10.1108/02635570610712573
of a ratio-format KPI that primarily relies on quantitative data (Hoehn, 2003). At thegroup and individual levels, there were many concepts such as motivational methodsbased on industrial psychologists and performance appraisals for salarystructure/workload analysis extended by human resource specialists, andpiece-rate/standard times determined by industrial engineers (Takala et al., 2006).
According to Sumanth (1985), Hoehn (2003) and Helo (2005), it is generally agreedthat productivity represents one of the major areas reflecting the term performance,especially for an organization or a production unit. Based on Sink and Tuttle (1989),there are seven performance criteria; i.e. profitability, productivity, quality of work life,innovation, quality, effectiveness, and efficiency. Productivity also represents anessential component of Harper’s performance measurement concept that includes unitcost, price, factor proportion, cost proportion, product mix, and input allocation (Harper,1984). It is vital for an organization to become continuously more productive in order tosustain its growth (Harper, 1984). Sumanth (1985) and Hoehn (2003) have pointed outthat becoming more productive implies better cost-competitiveness of and long-termprofitability for an organization. Specifically, productivity is simply the relationshipbetween the outputs (goods and services) generated from a system and the inputsprovided to create those outputs (Sink, 1985). Inputs include labor (human resources),capital (physical capital assets), energy, materials, and data that are brought into asystem. There have been many techniques for productivity measurement; e.g. theMFPMM, the Harper’s Performance Network, and Return on Assets (ROA).
According to Sink (1985), the MFPMM was developed by the AmericanProductivity Center in 1977 (nowadays known as the American Productivity andQuality Center) for measuring productivity and price recovery, and for explicitlyrelating these results with profitability at the organizational/functional levels. Itsprimary focus is on a manufacturing/production unit with tangible outputs and inputs.It is suitable for a process that is stable, implying not-so-often changes in productsbeing offered (Sink and Tuttle, 1989). Finally, the MFPMM can easily adapt the datafrom a typical accounting system. Figure 1 for its major components.
To elaborate more on the MFPMM, it is essential to highlight its other key features.For example, the MFPMM combines the concepts of the opportunity gain/loss, and thedynamic/static ratios in both single-, and partial-formats (Table I).
It is important to recognize at this point that measuring performance (i.e.productivity, profitability, etc.) needs to be suitable at a specific level that it is intended.Based on Sink and Tuttle (1989) and Singh et al. (2000), the analysis difficulties oftenarise when a plant manager has to rely on organizational-level or accountinginformation for monitoring technical-specific problems and for evaluating operational
Figure 1.Nine basic components for
the MFPMM
Change inoutput quantity
Change inProductivity
Change ininput quantity
Change inrevenues
Change inprofits
Change incosts
Change in outputprice (unitprice)
Change in price andcost ( price recovery)
Change in input cost(unit cost)
Productivity/performance
measurement
1273
performance. To further support this notion, Mandal (2000) states that the feedbackfrom improvement interventions that are initiated specifically within amanufacturing/production area is not usually visible at the organizational level. Asa result, it is essential that managers in charge of manufacturing or production beprovided with proper information.
Problems and objectivesThere were two critical problems facing a management team at the GPO productiondepartment. The first one was a lack of productivity and performance measurement,and subsequently specific information at the operational level for monitoring andevaluation. During the time of study, the accounting department had been mostlyresponsible for collecting production-related data, and for reporting performanceinformation. The second problem was a need to assess and evaluate whether theproduction department was ready to undertake its own performance-measurement and– analysis work. In the past, the production department had merely been informed ofits performance while the staffs from the accounting department routinely conductedperformance analysis and recommended possible improvement actions. The concernwas that if, specifically, the production department had its own measurement model,how well and effective performance information could have been utilized or analyzedthat would result in effective improvement interventions?
Year/category 2004 2005 Note
Revenue ($) 5,000 6,000Cost ($) 1,000 1,500Profit (revenue – cost) 4,000 4,500Profitability (revenue(cost)in a static format
5.0 4.0 Profit is not profitability. Profitdoes not necessarily implyprofitability. On the contrary,profitability results in profit
Profitability (revenue(cost) ina dynamic format
½ð6; 000 4 5; 000Þ4ð1; 500 4 1; 000Þ� or(1.2 4 1.5) ¼ 0.80
Rate of change in revenueVS rate of change in cost
Opportunity loss in $ (from aprofit-center point of view) –typically opportunity gain/lossnot included in the performanceor accounting report
21,500 If maintaining the ratio 5.0 fromyear 2004, the system should haveearned $7,500. Therefore, thesystem earned less than it shouldhave by $1,500 (based on7,500-6,000). However, thisnumber does not represent afinancial loss as shown by theaccounting system
Opportunity loss in $ (from acost-center point of view) – to beused in this project computation
2300 If maintaining the ratio 5.0 fromyear 2004, the system should haveconsumed $1,200. Therefore, thesystem consumed more than itshould have by $300 (based on1,500-1,200)
Source: Adapted from Sink (1985)
Table I.Demonstration of keyfeatures in the MFPMM
IMDS106,9
1274
Before deploying the MFPMM, a discussion session was held with the productiondepartment’s management team. This session focused on a selection of a suitablemeasurement technique. Included in this discussion was the possible use of value-addedlabor productivity and ROA. Even the term economic value-added (EVA) was brought up.Eventually, the MFPMM was selected due to the need to capture productivity informationas well as to relate this information with financial implications. Further, the information onopportunity gain/loss attracted a lot of attention. Since, it would be unwise to launch theMFPMM throughout the production department, it was initially suggested that oneproduction unit could be selected and tested. In this case, the tablet-production unit (Section3) was chosen. Therefore, the objectives of this project were to test whether the MFPMMwas a suitable tool to help initiate productivity/performance measurement for the GPO’sproduction department, and to assess and evaluate its performance-analysis capability.
It is important to note that the Section 3 unit did not experience low performance levels.It was selected because it represented a typical production unit. Further, the GPO was notin any immediate business danger. It is a state enterprise under the Ministry of PublicHealth. The GPO produces over 200 basic drugs in a complete range of dosage forms(tablets, capsules, creams, ointments, liquids, syrups, dry syrups, powders, and injections)for household consumption. Its overall goal has always been to produce pharmaceuticalproducts that support the public-health policy on the availability (implying large volumes)and the accessibility (implying low prices) of basic medicines for the poor. Given thisobligation, productivity/performance measurement became inevitable.
Approach and methodThere were several steps that had been taken for project completion. Some of the key tasksincluded the determination of the periods to be tested (Table II), the unit dimensions ofrequired data (more details later), and the actual computation (see Appendix 1 for acomprehensive MFPMM demonstration). The data of each period, for the Section 3, wasseparated into two major groups, namely outputs and inputs. It should be noted that thistask consumed a lot of time and was carefully executed. This was due to the need for amanagement team to understand that not all input factors could be captured or included inthe MFPMM. They were explained that the focus needed to be on the input factors thatrequired closed attention, and/or directly resulted in the outputs under study. Table III.The outputs to be examined were three package sizes of Paracetamol or “Para” tablet (10s,100s and 1,000s). The listed price (not considering retailers’ discounts) represented the unitprice to be charged. For the input side, the unit cost, based on the purchasing contracts,was used. For its quantity, the inputs were divided into five categories as follows:
(1) Labor was divided into four groups; i.e. manager, pharmacist, officers, andcasual workers. Each group was broken down into regular- and over-time.
(2) Materials included raw materials and packing materials.
(3) Energy was represented by water and electricity.
(4) Services were represented by the cost of maintenance and spare parts.
(5) Miscellaneous category included other costs such as welfare, office materials, etc.
ResultsFrom the departmental management team’s perspective, it was important to understandthe impacts from productivity and price-recovery on profitability. Essentially, the key
Productivity/performance
measurement
1275
concern was the required magnitude on the overall productivity improvement thatneeded to be achieved in order to make up for or offset the GPO inability to increase a unitprice of its products. Given the use of Minitab, the regression equation, based onthe columns 14, 15, and 16 – representing the overall results on productivity,price-recovery, and profitability was computed and can be demonstrated as follows. SeeAppendix 2:
Profitability ¼ 20:989 þ 1:00ðProductivityÞ þ 0:987ðPrice recoveryÞ
Given this finding (Figure 2 and Table IV), it was important that continuousproductivity improvement be given high priority. This implies the following issues.
Issue 1To maintain (not improving) the profitability level of 1.0, both the overall values ofproductivity and price recovery needed to be approximately 1.0. Given the circumstancefacing the GPO, the overall value of productivity needed to continuously achieve a levelhigher than 1.0 in order to compensate for the negative constant value of 0.989. This wasagain due to the GPO’s inflexibility and restriction on price-recovery.
Base period (period 1) Current or comparing period (period 2)
July 2002 August 2002August 2002 September 2002September 2002 October 2002October 2002 November 2002November 2002 December 2002December 2002 January 2003January 2003 February 2003February 2003 March 2003March 2003 April 2003April 2003 May 2003May 2003 June 2003June 2003 July 2003July 2003 August 2003August 2003 September 2003September 2003 October 2003October 2003 November 2003November 2003 December 2003December 2003 January 2004January 2004 February 2004February 2004 March 2004March 2004 April 2004April 2004 May 2004May 2004 June 2004June 2004 July 2004July 2004 August 2004August 2004 September 2004September 2004 October 2004October 2004 November 2004November 2004 December 2004
Table II.Base and current periodsfor the MFPMMimplementation
IMDS106,9
1276
Quantity Price Value
OutputsPara 10s Pack Baht/pack BahtPara 100s Bottle Baht/bottle BahtPara 1,000s Bottle Baht/bottle BahtInputsLaborManager Man-hr Baht/man-hr BahtPharmacist Man-hr Baht/man-hr BahtOfficers Man-hr Baht/man-hr BahtCasual workers Man-hr Baht/man-hr BahtMaterialsRaw materialParacetamol DC Kg Baht/kg BahtPacking materialsSticker 10s Each Baht/each BahtCarton No. 46 Each Baht/each BahtFoil Roll Baht/roll BahtBox Each Baht/each BahtSticker 100s Each Baht/each BahtCarton No. 49 Each Baht/each BahtPE-100s Bottle Baht/bottle BahtAl Foil Each Baht/each BahtSpongy 100s Each Baht/each BahtSticker 1,000s Kg Baht/kg BahtCarton no. 23 Each Baht/each BahtPE-1,000s Bottle Baht/bottle BahtSpongy 1,000s Kg Baht/kg BahtEnergyWater Unit Baht/unit BahtElectricity Unit Baht/unit BahtServicesMaintenance and spare parts Month Baht/month BahtMiscellaneous Month Baht/month Baht
Table III.Dimensional units for the
MFPMM application
Figure 2.Graphical presentation on
productivity,price-recovery and
profitability
0.80
0.85
0.90
0.95
1.00
1.05
1.10
1.15
1.20
Jul0
2-A
ug02
Aug
02-S
ep02
Sep0
2-O
ct02
Oct
02-N
ov02
Nov
02-D
ec02
Dec
02-J
an03
Jan0
3-Fe
b03
Feb0
3-M
ar03
Mar
03-A
pr03
Apr
03-M
ay03
May
03-J
un03
Jun0
3-Ju
l03
Jul0
3-A
ug03
Aug
03-S
ep03
Sep0
3-O
ct03
Oct
03-N
ov03
Nov
03-D
ec03
Dec
03-J
an04
Jan0
4-Fe
b04
Feb0
4-M
ar04
Mar
04-A
pr04
Apr
04-M
ay04
May
04-J
un04
Jun0
4-Ju
l04
Jul0
4-A
ug04
Aug
04-S
ep04
Sep0
4-O
ct04
Oct
04-N
ov04
Nov
04-D
ec04
Wei
ghed
Per
form
ance
ProductivityPrice RecoveryProfitability
Productivity/performance
measurement
1277
Issue 2The recent governmental policy for universal-health coverage placed the primaryresponsibility of producing affordable and basic medicines under the GPO. Therefore,productivity improvement probably represents the only viable alternative to achieve aprofitable status. It should be noted that a profitable operation for a state enterprise’sperspective does not represent the for-profit goal but reflects the ability to rely ononeself without financial assistance from the central government (while fulfilling itsmissions and goals).
Issue 3The opportunity loss created a lot of concern. The opportunity losses (not necessaryimplying actual financial losses – referred to from Table I) surprised most seniormanagers at the production department, especially from the productivity side (wheremost of improvement interventions from the organization were intended to focus). TheGPO had earlier attempted to improve its operations through inventory, delivery, andmaintenance. Several issues also came up during the review, including the possibilityof outsourcing work or of splitting production into two independent units. In regard tothe second issue, the first unit was for the so-called high-end customers whom would
Productivity Price recovery ProfitabilityValue Baht Value Baht Value Baht
July 2002-August 2002 0.9909 27,6594.32 1.0020 16,414.37 0.9928 260,179.95August 2002-September 2002 1.0021 18,324.52 1.0100 83,422.66 1.0122 101,747.18September 2002-October 2002 0.9689 225,5510.91 1.0921 702,267.08 1.0581 446,756.17October 2002-November 2002 1.0607 51,1374.19 1.0060 54,772.54 1.0671 566,146.72November 2002-December 2002 0.9497 2317,140.50 0.9798 2125,113.13 0.9304 2442,253.62December 2002-January 2003 1.0469 348,863.17 0.9891 281,021.77 1.0354 267,841.40January 2003-February 2003 1.0233 141,569.29 1.0094 56,527.99 1.0329 198,097.27February 2003-March 2003 1.0072 42,550.77 0.9468 2323,855.84 0.9537 2281,305.07March 2003-April 2003 1.0335 157,424.74 1.0082 39,057.81 1.0420 196,482.56April 2003-May 2003 0.9475 2310,361.91 1.0234 133,882.42 0.9697 2176,479.48May 2003-June 2003 0.9722 2200,184.11 1.0312 217,649.26 1.0025 17,465.15June 2003-July 2003 1.0020 13,459.08 0.9530 2341,545.79 0.9548 2328,086.71July 2003-August 2003 1.0548 377,551.22 1.0549 351,886.65 1.1127 729,437.87August 2003-September 2003 0.9624 2187,776.53 0.9605 2199,998.34 0.9244 2387,774.87September 2003-October 2003 0.9071 2374,743.29 1.0468 177,168.32 0.9496 2197,574.97October 2003-November 2003 1.1634 1,040,355.42 0.9539 2321,070.29 1.1098 719,285.13November 2003-December 2003 0.9052 2406,850.94 1.0856 337,892.78 0.9827 268,958.15December 2003-January 2004 1.0625 285,255.42 1.0185 76,974.67 1.0821 362,230.10January 2004-February 2004 0.9519 2255,173.00 0.9785 2119,195.30 0.9315 2374,368.30February 2004-March 2004 0.9842 285,540.81 1.0021 10,703.79 0.9863 274,837.01March 2004-April 2004 1.0290 167,544.27 0.9962 223,367.83 1.0251 144,176.44April 2004-May 2004 1.0106 73,638.74 0.9528 2345,968.01 0.9630 2272,329.27May 2004-June 2004 0.9893 287,047.65 0.9918 265,856.62 0.9813 2152,904.27June 2004-July 2004 1.0136 105,254.80 1.0152 118,323.54 1.0290 223,578.33July 2004-August 2004 0.9636 2201,995.54 0.9304 2414,418.58 0.8965 2616,414.12August 2004-September 2004 0.9410 2297,669.54 1.0984 451,439.08 1.0336 153,769.54September 2004-October 2004 1.0035 19,818.15 1.0110 62,454.46 1.0145 82,272.61October 2004-November 2004 1.0235 107,189.68 1.0349 156,166.46 1.0592 263,356.14November 2004-December 2004 1.0082 26,826.10 0.8173 2718,744.06 0.8240 2691,917.96
Table IV.Overall results onproductivity, pricerecovery, andprofitability
IMDS106,9
1278
be willing to purchase the medicines at the market price – more flexibility onprice-recovery. The second unit was mainly for the poor – less flexibility onprice-recovery. This is expected help better plan on the required magnitude ofproductivity improvement. However, these were not part of the project scope.
Issue 4The material input represented the most important input component (based onAppendix 1). Therefore, the tasks relating to price forecasting, contracting, productionplanning, inventory control, and exchange rate estimation would have to be examinedmore carefully. A plan was also being considered on experimenting with ane-procurement system. This could possibly result in a lower purchasing cost. Further,the issues of seasonal demand and its impacts on material management and productionwere discussed. However, due to the constant demand of basic medicines and theGPO’s main policy of primarily providing these medicines, the two issues were not theprimary concern when reviewing the above equation.
DiscussionThe follow-up review sessions took place after the results (i.e. the MFPMM outputs)had been demonstrated and analyzed. Generally, the use of MFPMM was perceived tobe useful and provided helpful information for policy- and decision-making. It helpedstimulate the discussion on potential improvement interventions and processimprovement (as previously described in the third and fourth implications,respectively). Simply put, the MFPMM by and large fulfilled the requirement setforth by the GPO and production department’s management on measuringproductivity and financially quantifying its impacts on the profit/loss. In addition,some of the departmental management members expressed the desire to illustrateinformation on opportunity gain/loss on a management report. This information wasbelieved to be quite useful for future monitoring and evaluation.
Based on these review sessions with the management team and line supervisors, thechallenges that the GPO would need to overcome in order to integrate the MFPMM canbe summarized as follows. The MFPMM practicality was the central issue of the debate.It was agreed that its sustainability and acceptance would depend on the ability of themanagement team to analyze performance information. However, the confidence oninformation interpretations, analysis skills, and knowledge on analysis tools (e.g. use ofthe Pareto Diagram to help portray the results from Columns 10 and 11, the regressionanalysis to show the different interrelationships among productivity, price-recovery,and profitability in terms of the time-lag effects, etc.) was not high, especially among linesupervisors. These supervisors presumably would be assigned the role of performance– information analysts. See Appendix 3 for other interrelationships from a regressionanalysis. Although having proper background, a lack of analysis experience and anissue of accountability (since an analysis decision would lead to improvementinterventions, and continuous measurement would eventually provide the feedback onthe effectiveness of these interventions) created a lot of anxiety among them. This wasunderstandable when the use of quantitative information suddenly became moreapparent in a department that previously relied on judgment/experience and analysisreports from the accounting department for planning. Moreover, due to its status as astate enterprise, the need to measure, analyze, and continuously improve its
Productivity/performance
measurement
1279
performance might not have been of great concern in the past. This situation hadrestricted the experience on analysis skills.
The next key issue was the evaluation of the database readiness for a future use ofthe MFPMM or other productivity/performance measurement techniques. Since, therelevant data had been recorded and stored at two locations (i.e. the accounting andproduction departments), it was necessary that a change would have to be made. It wasimportant that the data needed to be standardized in terms of output and inputclassifications, and to be uniformly maintained. Otherwise, the measurement effortwould only represent merely a monitoring effort (or lagging effort) without muchusefulness for analysis and improvement (implying forward-looking or leading effort).It should be noted that, for this experiment, a special effort had to be made in order toobtain necessary data for the MFPMM. The reason was that there was a time overlapbetween the two departments’ data collection frequencies. Some data were collected onthe daily basis while others were recorded on the weekly, monthly, and quarterly basis.
For the possibility of departmental-wide implementing the MFPMM, it wouldrequire additional preparations from the GPO. Based on the lessons learned and thelimitations from this experiment, the future preparations need to be made in twoaspects. The first one is on technical preparations. Some of the key steps that the GPOneeds to take for the MFPMM implementation are as follows:
. clarity on inputs to be captured;
. standardization of data to be collected – outputs and inputs;
. verification of data accuracy;
. assignment of a responsible team for these collection and verification tasks;
. identification of information-review frequency;
. developing a management report for analysis and decisions/actions; and
. recording such decisions for further follow-ups.
The second aspect was on managerial preparations. This would include primarily:. readiness assessment of the staffs to be assigned a task of conducting
performance analysis such as knowledge and experience on quality tools (e.g.Pareto and cause-and-effect diagrams); and
. assurance on skills and expertise of a management team on selecting appropriateproductivity/performance improvement techniques.
In addition, a clearer policy is likely needed on a lag-time effect from improvementinterventions. Unrealistic expectation of immediate improvement may discourage thecontinuous use of the MFPMM and other possible future performance measurementtechniques.
Finally, these difficulties confronting an effort on performance measurement at theGPO’s production department are typical and usual for any organization – not adiscouraging sign for the participating management team. Based on Hoehn (2003), thecareful consideration must be given to the scope and definition of what to be measuredand data to support such measures, database readiness, data accuracy, managementreview, accountability, and analysis for improvement. Furthermore, the ability tocontinuously manage the performance, with the focus on linking with analysis and
IMDS106,9
1280
improvement, is essential for sustaining and ensuring the success of performancemeasurement (Aneksitthisin, 2000; Saengchote et al., 2006). Otherwise, it is possiblethat performance measurement subsequently becomes another stand-alonemanagement tool. As a result, sustaining the use of performance measurementrequires a great deal of attention on both technical and managerial aspects.
Potential future workGiven the positive attitude toward the MFPMM usefulness, there are many potentialworks to be explored in order to ensure its continuity. One possibility is to further testthe MFPMM with other production sections that produce more medicine varieties thanthe Section 3. This can help further convince its applicability and implement-ability forthe entire production department. In addition, the departmental management team canclearly communicate or signify its intention to use performance measurement forimprovement purposes by benchmarking with other organizations such as OLIC(Thailand) or S.M. Pharmaceutical, especially in the areas of productivity. Relevantprocesses on material management such as inventory control, production planning,and procurement should be part of the future GPO’s benchmarking study.
To help analyze the cost drivers, the information report needs to contain the resultsfrom Columns 10 and 11 on the continuous basis. The suggestion on the ParetoDiagram during the experiment at the Section 3 appeared to be helpful for informationportrayal. In addition, they can examine other possibilities for the MFPMM’s long-termintegration. Rao (2006) provides a comprehensive demonstration on a spreadsheetapplication that resulted in a friendlier condition for a user. Furthermore, Rao andMiller (2004) suggest the possible use of an expert system to better utilize the MFPMM,especially for decision-making, and managing knowledge and lessons learned. Finally,it is important that a further analysis needs to be made into quantifying the impacts onproductivity, price-recovery, and profitability from other performance perspectivessuch as quality of working life and supplier development.
ConclusionThe project represented an effort to directly measure productivity/performance for theproduction department. Generally, the MFPMM received positive feedback on itsusefulness when tested at the Section 3. On the other hand, there are still many issuesthat need to be resolved before the possible MFPMM department-wide implementation.The ability to properly interpret and analyze productivity/performance informationneeds to be addressed. The more effective use of statistical tools has to be consideredwhen enhancing this ability. The data collection effort and the database managementalso needs to be strengthened.
References
Aneksitthisin, E. (2000), Implementation of the Multi-factor Productivity Measurement Model inthe Wood Industry: A Case Study at Vichit Parquet Ltd, Independent Study at KasetsartUniversity, Bangkok.
Duke, J. and Torres, V. (2005), “Multifactor productivity change in the air transportationindustry”, Monthly Labor Review, (sponsored by Bureau of Labor Statistics, Department ofLabor, USA), March.
Harper, J. (1984), Measuring Business Performance, Institute of Manpower Studies, GowerPublishing, Aldershot.
Productivity/performance
measurement
1281
Helo, P. (2005), “Managing agility and productivity in the electronics industry”, IndustrialManagement & Data Systems, Vol. 104 No. 7, pp. 567-77.
Hoehn, W.K. (2003), “Managing organizational performance: linking the balanced scored to aprocess improvement technique”, Proceeding of the 4th Annual International Symposiumin Industrial Engineering on the Performance-based Management, Kasetsart University,Bangkok, pp. 1-12.
Kaplan, R.S. and Norton, D.P. (2004),TheStrategyMaps, Harvard Business School Press, Boston, MA.
Kaydos, W. (1991), Measuring, Managing, and Maximizing Performance, Productivity Press,Atlanta, GA.
Mandal, P. (2000), “Inter-functional spread of quality in manufacturing”, Industrial Management& Data Systems, Vol. 100 No. 3, pp. 135-40.
Meyer, P.B. and Harper, M.J. (2005), “Preliminary estimates of multifactor productivity growth”,Monthly Labor Review, (sponsored by Bureau of Labor Statistics, Department of Labor,USA), June.
Rao, M.P. (2006), “A performance measurement system using a profit-linked multi-factormeasurement model”, Industrial Management & Data Systems, Vol. 106 No. 3, pp. 362-79.
Rao, M.P. and Miller, D.M. (2004), “Expert systems applications for productivity analysis”,Industrial Management & Data Systems, Vol. 104 No. 9, pp. 776-85.
Saengchote, P., Phusavat, K. and Takala, J. (2006), “Applying the performance network conceptfor productivity measurement and target setting: case discussion on an electricmanufacturing plant in Thailand”, Proceeding of the 14th International Working Seminaron Production Economics, pp. 307-24.
Singh, H., Motwani, J. and Kumar, A. (2000), “A review and analysis of the state-of-the-artresearch on productivity measurement”, Industrial Management & Data Systems, Vol. 105No. 5, pp. 234-41.
Sink, D.S. (1985), Productivity Management: Planning, Measurement and Evaluation, Control andImprovement, Wiley, New York, NY.
Sink, D.S. and Tuttle, T.C. (1989), Planning and Measurement in Your Organization of theFuture, IE Press, Norcross, GA.
Sumanth, D.J. (1985), Productivity Engineering and Management, McGraw-Hill, New York, NY.
Takala, J., Suwansaranyu, U. and Phusavat, K. (2006), “A proposed white-collar workforceperformance measurement framework”, Industrial Management & Data Systems, Vol. 106No. 5, pp. 644-62.
Tsai, W.H., Hsu, P.Y., Cheng, J.M.S. and Chen, Y.W. (2006), “An AHP approach to assessing therelative importance weights of ERP performance measures”, International Journal ofManagement and Enterprise Development, Vol. 3 No. 4, pp. 351-75.
Further reading
Dixon, J.R., Nanni, A.J. and Vollmann, T.E. (1990), The New Performance Challenges: MeasuringOperations of World-Class Competitions, Down Jones-Irwin, New York, NY.
Green, K.W., Medlin, B. and Whitten, D. (2004), “Developing optimism to improve performance:an approach for the manufacturing sector”, Industrial Management & Data Systems,Vol. 104 No. 5, pp. 106-14.
Helo, P. and Szekely, B. (2004), “Logistics information systems”, Industrial Management & DataSystems, Vol. 105 No. 2, pp. 5-18.
Mistry, J. (2006), “Differential impacts of information technology on cost and revenue driverrelationships in banking”, IndustrialManagement &Data Systems, Vol. 106 No. 3, pp. 327-44.
IMDS106,9
1282
Appendix 1. MFPMM demonstration
Figure A1.
Complete MFPMM
results
Productivity/performance
measurement
1283
Based on Sink (1985), the formulas for the MFPMM can be shown as follows.Column 1: Representing the quantities of outputs produced and/or sold and the quantities of
input consumed in order to produce those outputs for period 1 (Qi1). The term “i” representsdifferent categories of outputs and inputs. For example, the input factor consists of labor,material, etc.
Column 2: Representing the outputs’ unit prices and the inputs’ unit cost in period 1 (Pi1).Column 3: Reflecting the value (quantity £ unit price or unit cost) of period 1:
Column 3 ¼ Column 1 £ Column 2
Column 4: Representing the quantities of outputs produced and/or sold and the quantities ofinput consumed in order to produce those outputs for period 2 (Qi2).
Column 5: Representing the outputs’ unit prices and the inputs’ unit cost in period 1 (Pi2).Column 6: Reflecting the value (quantity £ unit price or unit cost) of period 2:
Column 6 ¼ Column 4 £ Column 5
Column 7: Price-weighted and base period price indexed changes – quantities (both unit priceand cost remained constant at period 1):
Xni¼1
ðQi2ÞðPi1Þ
Xni¼1
ðQi1ÞðPi1Þ
Column 8: Quantity-weighted and current period indexed changes – unit prices and costs (bothoutput and input quantities remained constant at period 2):
Xni¼1
ðQi2ÞðPi2Þ
Xni¼1
ðQi2ÞðPi1Þ
Column 9: Examining the impact of changes in price and quantity from period 1 to 2 – quantity(for outputs and inputs) and unit price/cost not held constant:
Xni¼1
ðQi2ÞðPi2Þ
Xni¼1
ðQi1ÞðPi1Þ
Note: There is a need to separate the term “Qi” or quantity into “Oi” reflecting specifically theoutputs produced, and “Ii” representing only the inputs consumed. This is necessary since thecomputation for the remaining columns will focus on both the output and input sidessimultaneously. In addition, it is often that the input side be further divided into differentcategories and sub-categories, respectively. On the contrary, for the output side, it seldomcontains at the sub-category level. For example, the labor input can have management,supervisors, and operation categories. The operational-labor category can be separated intoassembly- line, warehouse, and quality inspection sub-categories. As a result, the term “Ii” is tobe denoted as “Iij.” It should be noted that it is possible that each output category can have
IMDS106,9
1284
several corresponding sub-categories. However, the formulas, provided by Sink (1985), do notreflect this need for the output side.
Column10: Examining the cost-to-revenue for period 1:
I ij1Xni¼1
ðOi1ÞðPi1Þ
Column 11: Examining the cost-to-revenue for period 2:
I ij2Xni¼1
ðOi2ÞðPi2Þ
Column 12: Examining the output-to-input ratios for period 1:
Xni¼1
ðOi1ÞðPi1Þ
ðI ij1ÞðPij1Þ
Column 13: Examining the output-to-input ratios for period 2:
Xni¼1
ðOi2ÞðPi1Þ
ðI ij2ÞðPij1Þ
Column 14: Reflecting price-weighted productivity indexes:
Xni¼1
ðOi2ÞðPi1Þ=Xni¼1
ðOi1ÞðPi1Þ
)
ðI ij2ÞðPi1Þ=ðI ij1ÞðPi1Þ
)Rate of change on input quantity consumed
Rate of change on output quantity produced
Column 15: Representing quantity – weighted price recovery indexes:
Xni¼1
ðOi2ÞðPi2Þ=Xni¼1
ðOi2ÞðPi1Þ
)
ðI ij2ÞðPi2Þ=ðI ij2ÞðPi1Þ
)Rate of change on input’s unit cost
Rate of change on output’s unit price
Column 16: Depicting profitability indexes, reflecting the rates of change on both the quantityproduced/consumed and unit price/cost:
Xni¼1
ðOi2ÞðPi2Þ=Xni¼1
ðOi1ÞðPi1Þ
�
ðI ij2ÞðPi2Þ=ðI ij1ÞðPi1Þ
�Rate of change on input value
Rate of change on output value
Column 17: Indicating the impacts (opportunity gain/loss) from a productivity change:
Productivity/performance
measurement
1285
ðI ij1ÞðPi1Þ� �
Xni¼1
ðQi2ÞðPi1Þ
" #P
ðQi1ÞðPi1Þ
0BBBB@
1CCCCA2
ðI ij2ÞðPi1Þ
ðI ij1ÞðPi1Þ
� �266664
377775
Column 18: Indicating the impact from a price recovery change:
Column 18 ¼ Column 19 2 Column 17
Column 19: Indicating the total impact on profits from productivity/price recovery:
ðI ij1ÞðPi1Þ� �
Xni¼1
ðQi2ÞðPi2Þ
" #P
ðQi1ÞðPi1Þ
0BBBB@
1CCCCA2
ðI ij2ÞðPi2Þ
ðI ij1ÞðPi1Þ
� �266664
377775
Appendix 2. Computation of the regression analysis for the interrelationshipsamong profitability, productivity, and price recoveryColumn16 ¼ 20.989 þ 1.00 Column 14 þ 0.987 Column 15 or Profitability ¼ 20.989 þ 1.00(Productivity) þ 0.987 (price recovery)
Analysis of variance
Predictor Coefficient SE coefficient T P
Constant 20.98946 0.01520 265.08 0.000Column 14 1.00203 0.00978 102.45 0.000Column 15 0.986690 0.00930 106.07 0.000
Notes: S ¼ 0.002607 R 2 ¼ 99.8 per cent, R 2 (adjusted) ¼ 99.8 per centTable AI.
Source DF SS MS F P
Regression 2 0.116547 0.058273 8572.15 0.000Residual error 26 0.000177 0.000007Total 28 0.116724Table AII.
Source DF Seq. SS
Column 14 1 0.040064Column 15 1 0.076483Table AIII.
IMDS106,9
1286
Unusual observations
Appendix 3. Other results from the regression analysisThis appendix demonstrates the assistance that was provided during the review sessions on theMFPMM results with the production department’s management team. During these reviews,several suggestions were made to help better understand, analyze, and interpret performanceinformation derived from the MFPMM. For example, different interrelationship conditionsneeded to be examined, especially between productivity and profitability. Interestingly, for thisexample, only two conditions were acceptable for further analysis, given the initial experiment atthe Section 3. Based on these results, productivity immediately impacts the level of profitability.For the results from the time-lag factor, the one-month aspect was more helpful than other threetime-lag conditions for analyzing the overall impact of productivity on profitability. In this case,it can simply be concluded that productivity has immediate impacts on profitability. Therefore,the management team is advised to be urgently alarmed when there was a decline inproductivity.
Corresponding authorKongkiti Phusavat can be contacted at: fengkkp@ku.ac.th
Obs Col. 14 Col. 16 Fit SE Fit Residual St. Residual
16 1.16 1.1098 1.11750 0.00162 20.00770 23.77RX17 0.91 0.9827 0.98873 0.00115 20.00603 22.57R29 1.01 0.8240 0.82720 0.00175 20.00320 21.66 X
Notes: “R” denotes an observation with a large standardized residual; “X” denotes an observationwhose X value gives it large influence Table AIV.
Description: profitability is denotedas “Y.” Productivity is denoted as “X.” Regression equation results
R 2
(adjusted)(percent)
p-value(a ¼ 0.05)
Condition 1: without the time factor Y ¼ 0.276 þ 0.723(X) 31.9 0.001Condition 2: with the time-lag factorof one month Y ¼ 1.56 2 0.563(X)t21 17.7 0.015Condition 3: with the time-lag factorof two months Y ¼ 0.870 þ 0.129(X)t22 0.0 0.606Condition 4: with the time-lag factorof three months Y ¼ 1.01 2 0.017(X)t23 0.0 0.945 Table AV.
Productivity/performance
measurement
1287
To purchase reprints of this article please e-mail: reprints@emeraldinsight.comOr visit our web site for further details: www.emeraldinsight.com/reprints