"An urgent order. A customer demanding a shorter delivery date. And us, staring at a system that gives us no clear answer on whether we have the capacity to say yes or no. We have been here before. We accepted the order under pressure. We made the promise. And we paid the price: delayed existing orders, unhappy customers, and penalties we did not budget for. Without a reliable production scheduling tool, every urgent order is not an opportunity. It is a risk we are taking blindfolded."     This is not an isolated complaint. Many pharma companies have invested millions in production scheduling software to solve exactly this problem, yet most have either shelved these tools entirely or confined them to generating broad monthly production numbers. Anecdotal evidence suggests that large pharma companies, despite the investment, have gone back to MS Excel.   This article explores why. <h2>Why is software needed at all?</h2>Pharmaceutical plants, such as the one making oral solids, operate with multiple machines at each work centre: Granulation, Blending, Compression, Coating, and Packing. The product portfolio is vast - every unique combination of molecule, pack size, and country of sale is treated as a distinct product, each following a designated route through the plant. These routes crisscross, creating a web of interdependencies (Diagram 1). Processing times vary across machines and products. Campaign formation requirements add further constraints, while changeovers consume significant machine capacity. In this environment, building a schedule manually can take planners several days.<figure class="image"><img style="aspect-ratio:1210/519;" src="https://web-assets.vectorconsulting.in/storage/editor_uploads//kiran_1780384284.png" width="1210" height="519"></figure>Diagram 1: Production set-up and interdependencies in a pharma plant <h2>Why does adoption remain low?</h2>The failure is not in the software. It is in how planning actually works on the plant floor.   Plants prepare a monthly production plan and derive monthly output targets from it. These targets become the benchmark against which management evaluates performance and fixes accountability.   To hit these targets, the plant needs sufficient work-in-progress (WIP) organised in campaigns across all production routes. This requires dispensing to be secured first. The result is a dispensing plan that is largely frozen for the month.   The frozen dispensing plan is non-negotiable for two reasons: <ol><li>It is designed around campaigns that minimise changeovers.</li><li>It is structured to maintain consistent WIP across all routes. </li></ol> Any deviation erodes volume. Any WIP imbalance brings production to a halt on routes where WIP is low. Either way, the monthly target is at risk. All the problems begin here.  Reality never matches the plan. Raw material delays, Quality Control (QC) failures, BMR/BPR issues, operator absenteeism, and machine breakdowns are not exceptions; they are the norm. When disruption originates on the plant floor, dispensing as per the frozen plan causes WIP to pile up in front of the affected machine.    When disruption originates upstream, through a raw material delay or QC failure, the plant pulls forward the next scheduled batches to keep granulation running. But because every product follows a designated route, this does not translate into utilisation downstream. WIP accumulates in front of some machines while others remain starved. The imbalance inflates inventory, masks idle capacity, and drags overall output down.   When output falls short of target, compression and packing sections selectively prioritise batches that maximise production numbers - larger or simpler batches - at the expense of delivery dates. Many times, the delayed orders often attract penalties and air freight costs. To recover, the plant runs overtime and breaks campaigns to accommodate urgent batches.    More changeovers mean more downtime. Output falls further. A vicious cycle takes hold.   It is not uncommon for planners to remain in constant firefighting mode. When customers request urgent orders, there is no bandwidth left to assess short-term capacity and no credible basis on which to commit.  This is precisely the problem scheduling software is meant to solve. Two reasons explain why it often fails:<ol><li><h5>Revised plans lower the monthly target</h5>When scheduling software runs against current reality, the revised plan almost always reflects a lower output target. In most organisations, this is unacceptable. The more frequently the software is run, the more the target appears to erode. A lower target feels like an admission of further failure. Management does not accept it. Therefore, the core output of the software - a revised schedule - goes unused. The original schedule is preserved, and the burden falls on the plant to somehow achieve the number by month-end.  </li><li><h5>Real-time WIP data is unavailable</h5>Since the revised schedule is not used, there is little incentive to maintain live WIP tracking. Usually, WIP data is fed into the scheduler manually at the start of the month. Repeating this exercise mid-month demands effort and planner bandwidth that is already exhausted by firefighting. In OSD (Oral Solid Dose) plants, WIP spans 25-40 days from dispensing to packing. Without current WIP status, the plant model inside the software is always outdated. Any urgent-order assessment drawn from it becomes unreliable. </li></ol>Both failures trace back to the same root cause: the fixed monthly target paradigm. Management operates on a ‘fix and hold’ logic. The software is built for a dynamic world. The two are incompatible. <h2>The way forward</h2>Plants need to stop treating the monthly plan as a fixed target and start treating it as the latest best estimate of what the plant can achieve. The trouble begins when estimates are converted into commitments in a variable environment and then defended rigidly against reality. A  variable environment needs high-frequency re-estimation, not frozen commitments. Once that shift in thinking happens, scheduling software can function as intended: continuously adapting the schedule based on actual capacity and raw material status, dynamically preponing and postponing batches, while still honouring customer commitments.    The dispensing plan becomes dynamic, ensuring balanced WIP across every route. This, in turn, mandates real-time WIP visibility. As the month progresses and the plan diverges from actual execution, that divergence itself becomes the trigger for planners to rerun the scheduling engine and publish the latest execution status.   For customer delivery commitments to remain reliable despite dynamic execution, the commitment engine must operate separately from the execution engine. Commitments to customers should be made conservatively - planned below rated capacity - with manpower availability factored in. This built-in buffer protects delivery dates from the outset.    This separation resolves a conflict that has no solution within a single planning model. A conservative plan protects delivery reliability but appears to concede capacity. An aggressive plan minimises apparent capacity loss but pushes orders out whenever variability hits. When commitment and execution share the same plan, management is forced to choose one at the expense of the other.   Separating the two engines removes that trade-off. The commitment engine quotes dates with a buffer. The execution engine then works to beat those dates by dynamically deploying capacity as it becomes available in the short term. Reliability and utilisation are no longer in competition.  In the new planning model, instead of relying on a rolling monthly demand model, where one month is fixed and future months remain variable, often stretching 3-9 months out, sales should stack orders against a defined capacity file over a limited horizon. This prevents highly uncertain future demand from influencing near-term sourcing and execution decisions.   With a live execution model in place, idle capacity becomes visible within the schedule. That visibility is what enables an informed response to every urgent order possible.<h2> Necessary conditions for successful deployment</h2>Once the management paradigm shifts, two tactical conditions must also be in place.<ol><li><h5>Real-time WIP data: </h5>Given that OSD plants carry 25-40 days of WIP, the status of every batch in process must be known at the time of every scheduling run. Without this, the plan begins diverging from reality from day one.<h5> </h5></li><li><h5>Plan stability for at least seven days:</h5>A plan that cannot hold for a week has little practical value for the production team. Achieving this requires perpetual raw material availability, QC-approved raw material inventory sufficient for 15-20 days at all times, managed through an intelligent mix of stock- and demand-based ordering.   </li></ol><h5>Expected results*</h5>Plants that have adopted the suggested dynamic scheduling approach have reported:<figure class="image"><img style="aspect-ratio:2085/531;" src="https://web-assets.vectorconsulting.in/storage/editor_uploads//expected-result_1780385402.png" width="2085" height="531"></figure>(*Typical implementation results reported by Vector Consulting Group in pharma operations)<h2> Closing</h2>The pharmaceutical industry has spent millions on production scheduling software. Adoption remains low not because the tools are inadequate, but because the management paradigm into which they are deployed is incompatible with how they function. A fixed monthly target, used as an accountability instrument, makes dynamic rescheduling organisationally unacceptable. The real change required is not in the software. It is in  the mindware.

Key Takeaways

Pharma scheduling software fails because rigid monthly targets clash with plant-floor disruptions. Vector Consulting Group fixes this with a dynamic planning model, boosting output by 25% and securing over 95% on-time delivery.

Production scheduling software: Why they mostly fail in pharma environments

Production scheduling software: Why they mostly fail in pharma environments

Download PDF

Discover the root causes of planning failures and how a dynamic scheduling model balances WIP, protects delivery dates, and cuts costs in pharma plants.